# Compare models

- **URL canônica:** https://meirinho.com.br/en/ia/modelos/comparar/
- **Idioma:** en
- **Atualizado:** 2026-11-17

---
AI models

# Compare models

Pick a benchmark that already has a score, filter, and mark up to four models.

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/)

1. Pick a benchmark that already has a score.
2. Filter by name, lab or weights.
3. Mark up to four models.
4. Token price does not close the cost of the finished job.

## Benchmarks with a score

| Case | What to measure | Reference model | Open alternative | Source |
| --- | --- | --- | --- | --- |
| [Text](#gpqa-diamond) | GPQA Diamond | [GPT-6.2 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-2-astra/) 96.8% Epoch AI | [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) 94.2% Epoch AI | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) |
| [Text](#otis-aime) | OTIS Mock AIME | [Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/claude-opus-5-5/) 100% Epoch AI | [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) 100% Epoch AI | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) |
| [Text](#frontiermath) | FrontierMath (tiers 1 to 3) | [Claude Fable 5.2](https://meirinho.com.br/en/ia/modelos/claude-fable-5-2/) 95.1% Epoch AI | [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) 78.2% Epoch AI | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) |
| [Text](#simpleqa-verified) | SimpleQA Verified | [Claude Fable 5.2](https://meirinho.com.br/en/ia/modelos/claude-fable-5-2/) 94% Epoch AI | [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) 62.2% Epoch AI | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) |
| [Code](#swe-bench-verified) | SWE-bench Verified | [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) 83.5% Epoch AI | [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) 78.7% Epoch AI | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) |
| [Text](#or-gpqa-diamond) | GPQA Diamond (OpenRouter) | [Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/gemini-4-argon/) 96.5% OpenRouter Benchmarks | [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) 93.9% OpenRouter Benchmarks | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) |
| [Agents](#tau-bench-airline) | τ-bench Verified (airline) | [Claude Fable 5](https://meirinho.com.br/en/ia/modelos/claude-fable-5/) 82% OpenRouter Benchmarks | [DeepSeek V4 Flash 0423](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash/) 81.7% OpenRouter Benchmarks | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) |

## [GPQA Diamond](https://meirinho.com.br/en/ia/modelos/benchmark/gpqa-diamond/)

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/texto/)

| Model | % | Lab | Weights | Source | Compare models |
| --- | --- | --- | --- | --- | --- |
| [GPT-6.2 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-2-astra/) | 96.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-6.2 Astra |
| [Gemini 4 Carbon](https://meirinho.com.br/en/ia/modelos/gemini-4-carbon/) | 96.5% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 4 Carbon |
| [Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/gemini-4-argon/) | 96.4% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 4 Argon |
| [GPT-6 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-astra/) | 95.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-6 Astra |
| [Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5-5/) | 95.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Sonnet 5.5 |
| [GPT-6.1 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-1-sol/) | 95.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-6.1 Sol |
| [Gemini 3.8 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-8-flash/) | 95.4% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3.8 Flash |
| [Gemini 3.7 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-7-flash/) | 94.8% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3.7 Flash |
| [GPT-5.4 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-4-pro/) | 94.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.4 Pro |
| [Gemini 3.1 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-1-pro/) | 94.4% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3.1 Pro |
| [GPT-6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-sol/) | 94.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-6 Sol |
| [GPT-6.1 Terra](https://meirinho.com.br/en/ia/modelos/gpt-6-1-terra/) | 94.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-6.1 Terra |
| [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) | 94.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Kimi K3.1 Pro |
| [Gemini 3.6 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-6-flash/) | 94.1% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3.6 Flash |
| [GPT-5.5](https://meirinho.com.br/en/ia/modelos/gpt-5-5/) | 94% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.5 |
| [Grok 4.6](https://meirinho.com.br/en/ia/modelos/grok-4-6/) | 94% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok 4.6 |
| [Claude Fable 5.2](https://meirinho.com.br/en/ia/modelos/claude-fable-5-2/) | 93.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Fable 5.2 |
| [Claude Opus 5](https://meirinho.com.br/en/ia/modelos/claude-opus-5/) | 93.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 5 |
| [GPT-5.5 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-5-pro/) | 93.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.5 Pro |
| [GPT-5.6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol/) | 93.5% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.6 Sol |
| [Grok 4.5](https://meirinho.com.br/en/ia/modelos/grok-4-5/) | 93.4% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok 4.5 |
| [GPT-5.4](https://meirinho.com.br/en/ia/modelos/gpt-5-4/) | 93.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.4 |
| [GPT-5.6 Terra](https://meirinho.com.br/en/ia/modelos/gpt-5-6-terra/) | 93.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.6 Terra |
| [Kimi K3](https://meirinho.com.br/en/ia/modelos/kimi-k3/) | 93.1% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Kimi K3 |
| [Doubao Seed Evolving (Nov 2026)](https://meirinho.com.br/en/ia/modelos/doubao-seed-evolving-nov-2026/) | 93.1% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Doubao Seed Evolving (Nov 2026) |
| [Gemini 3.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash/) | 92.8% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3.5 Flash |
| [Grok 4.7](https://meirinho.com.br/en/ia/modelos/grok-4-7/) | 92.7% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok 4.7 |
| [Gemini 3 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-pro/) | 92.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3 Pro |
| [Qwen3.8 Max](https://meirinho.com.br/en/ia/modelos/qwen3-8-max/) | 92.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.8 Max |
| [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) | 91.9% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GLM-5.2 |
| [GPT-5.6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna/) | 91.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.6 Luna |
| [GPT-5.2](https://meirinho.com.br/en/ia/modelos/gpt-5-2/) | 91.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.2 |
| [Claude Opus 4.8](https://meirinho.com.br/en/ia/modelos/claude-opus-4-8/) | 91% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 4.8 |
| [DeepSeek V4 Flash 0731](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash-0731/) | 91% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek V4 Flash 0731 |
| [DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/deepseek-v4-pro/) | 90.9% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek V4 Pro |
| [GLM-5.3](https://meirinho.com.br/en/ia/modelos/glm-5-3/) | 90.9% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GLM-5.3 |
| [MiniMax M3](https://meirinho.com.br/en/ia/modelos/minimax-m3/) | 90.9% Epoch AI | MiniMax | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | MiniMax M3 |
| [Qwen3.7 Max](https://meirinho.com.br/en/ia/modelos/qwen3-7-max/) | 90.9% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.7 Max |
| [Kimi K2.6](https://meirinho.com.br/en/ia/modelos/kimi-k2-6/) | 90.8% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Kimi K2.6 |
| [Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/claude-opus-5-5/) | 90.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 5.5 |
| [GPT-6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-6-luna/) | 90.5% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-6 Luna |
| [Claude Opus 4.6](https://meirinho.com.br/en/ia/modelos/claude-opus-4-6/) | 90.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 4.6 |
| [Claude Sonnet 5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5/) | 90.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Sonnet 5 |
| [GLM-5.3-Flash](https://meirinho.com.br/en/ia/modelos/glm-5-3-flash/) | 90.2% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GLM-5.3-Flash |
| [Mistral Large 4](https://meirinho.com.br/en/ia/modelos/mistral-large-4/) | 90.2% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Large 4 |
| [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) | 90.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 4.7 |
| [GLM 5.1](https://meirinho.com.br/en/ia/modelos/glm-5-1/) | 89.9% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GLM 5.1 |
| [DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/deepseek-v4-1-flash/) | 89.8% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek V4.1 Flash |
| [Muse Spark](https://meirinho.com.br/en/ia/modelos/muse-spark/) | 89.8% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Muse Spark |
| [Claude Haiku 5.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-5-5/) | 89.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Haiku 5.5 |
| [Doubao Seed 2.1 Turbo](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-1-turbo/) | 89.6% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Doubao Seed 2.1 Turbo |
| [Gemini 3 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-flash/) | 89.4% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3 Flash |
| [Grok 4.20](https://meirinho.com.br/en/ia/modelos/grok-4-20/) | 89.3% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok 4.20 |
| [Doubao Seed 2.1 Pro](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-1-pro/) | 89.3% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Doubao Seed 2.1 Pro |
| [Grok 4.3 Beta](https://meirinho.com.br/en/ia/modelos/grok-4-3-beta/) | 88.8% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok 4.3 Beta |
| [Muse Spark Open](https://meirinho.com.br/en/ia/modelos/muse-spark-open/) | 88.6% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Muse Spark Open |
| [Qwen3.6 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-6-plus/) | 88.4% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.6 Plus |
| [Doubao-Seed-2.0 Pro](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-0-pro/) | 88.2% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Doubao-Seed-2.0 Pro |
| [Doubao Seed 2.0 Lite (Apr 2026)](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-0-lite-apr-2026/) | 88.1% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Doubao Seed 2.0 Lite (Apr 2026) |
| [Kimi K2.7 Code](https://meirinho.com.br/en/ia/modelos/kimi-k2-7-code/) | 87.9% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Kimi K2.7 Code |
| [Qwen3.7 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-7-plus/) | 87.9% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.7 Plus |
| [GLM 5](https://meirinho.com.br/en/ia/modelos/glm-5/) | 87.8% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GLM 5 |
| [Kimi K2.5](https://meirinho.com.br/en/ia/modelos/kimi-k2-5/) | 87.6% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Kimi K2.5 |
| [GPT-5.1](https://meirinho.com.br/en/ia/modelos/gpt-5-1/) | 87.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.1 |
| [Claude Sonnet 4.6](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-6/) | 87.4% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Sonnet 4.6 |
| [Qwen 3.6 Max (Preview)](https://meirinho.com.br/en/ia/modelos/qwen-3-6-max-preview/) | 87.4% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen 3.6 Max (Preview) |
| [Grok 4](https://meirinho.com.br/en/ia/modelos/grok-4/) | 87% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok 4 |
| [GPT-5.4 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-4-mini/) | 86.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.4 Mini |
| [Qwen3.5 397B A17B](https://meirinho.com.br/en/ia/modelos/qwen3-5-397b-a17b/) | 86.4% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.5 397B A17B |
| [GPT-5](https://meirinho.com.br/en/ia/modelos/gpt-5/) | 86.2% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5 |
| [Claude Opus 4.5](https://meirinho.com.br/en/ia/modelos/claude-opus-4-5/) | 86% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 4.5 |
| [Claude Fable 5](https://meirinho.com.br/en/ia/modelos/claude-fable-5/) | 85.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Fable 5 |
| [Qwen3.6 27B](https://meirinho.com.br/en/ia/modelos/qwen3-6-27b/) | 85.9% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.6 27B |
| [Nemotron 3 Ultra](https://meirinho.com.br/en/ia/modelos/nemotron-3-ultra/) | 85.4% Epoch AI | NVIDIA | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Nemotron 3 Ultra |
| [Gemini 2.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-2-5-pro/) | 85.3% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 2.5 Pro |
| [Qwen3.5 Plus 2026-02-15](https://meirinho.com.br/en/ia/modelos/qwen3-5-plus-02-15/) | 84.8% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.5 Plus 2026-02-15 |
| [Qwen3.6 35B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-6-35b-a3b/) | 84.8% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.6 35B A3B |
| [Kimi K2 Thinking](https://meirinho.com.br/en/ia/modelos/kimi-k2-thinking/) | 84.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Kimi K2 Thinking |
| [Qwen3.5-35B-A3B](https://meirinho.com.br/en/ia/modelos/qwen3-5-35b-a3b/) | 83.5% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.5-35B-A3B |
| [DeepSeek V3.2](https://meirinho.com.br/en/ia/modelos/deepseek-v3-2/) | 83.4% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek V3.2 |
| [Gemini 3.5 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash-lite/) | 83.3% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3.5 Flash Lite |
| [Qwen3.6 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-6-flash/) | 83.3% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.6 Flash |
| [GLM 4.7](https://meirinho.com.br/en/ia/modelos/glm-4-7/) | 83.3% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GLM 4.7 |
| [GPT-5.5 Instant](https://meirinho.com.br/en/ia/modelos/gpt-5-5-instant/) | 82.5% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.5 Instant |
| [Claude Sonnet 4.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-5/) | 82.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Sonnet 4.5 |
| [Qwen3.5-Flash](https://meirinho.com.br/en/ia/modelos/qwen3-5-flash-02-23/) | 82.3% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.5-Flash |
| [Qwen3.7 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-7-flash/) | 82.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.7 Flash |
| [Gemini 3.1 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-1-flash-lite/) | 81.8% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 3.1 Flash Lite |
| [o3](https://meirinho.com.br/en/ia/modelos/o3/) | 81.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | o3 |
| [Qwen3-235B-A22B](https://meirinho.com.br/en/ia/modelos/qwen3-235b-a22b/) | 80.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3-235B-A22B |
| [Claude 3.7 Sonnet](https://meirinho.com.br/en/ia/modelos/claude-3-7-sonnet/) | 79.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 3.7 Sonnet |
| [o4-mini](https://meirinho.com.br/en/ia/modelos/o4-mini/) | 79.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | o4-mini |
| [Claude Sonnet 4](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4/) | 79.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Sonnet 4 |
| [Qwen3.5-9B](https://meirinho.com.br/en/ia/modelos/qwen3-5-9b/) | 79% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3.5-9B |
| [GPT-5.4 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-4-nano/) | 78.5% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5.4 Nano |
| [Doubao-Seed-1.6](https://meirinho.com.br/en/ia/modelos/doubao-seed-1-6/) | 77.6% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Doubao-Seed-1.6 |
| [Claude Opus 4.1](https://meirinho.com.br/en/ia/modelos/claude-opus-4-1/) | 77.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 4.1 |
| [o3-mini](https://meirinho.com.br/en/ia/modelos/o3-mini/) | 77% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | o3-mini |
| [o1](https://meirinho.com.br/en/ia/modelos/o1/) | 76.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | o1 |
| [DeepSeek R1](https://meirinho.com.br/en/ia/modelos/deepseek-r1/) | 76.3% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek R1 |
| [Claude Opus 4](https://meirinho.com.br/en/ia/modelos/claude-opus-4/) | 76.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Opus 4 |
| [Grok-3 mini](https://meirinho.com.br/en/ia/modelos/grok-3-mini/) | 76.3% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok-3 mini |
| [gpt-oss-120b](https://meirinho.com.br/en/ia/modelos/gpt-oss-120b/) | 75.8% Epoch AI | OpenAI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | gpt-oss-120b |
| [Grok 3](https://meirinho.com.br/en/ia/modelos/grok-3/) | 75.8% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok 3 |
| [GPT-5 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-mini/) | 75% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5 Mini |
| [Gemma 4 26B A4B](https://meirinho.com.br/en/ia/modelos/gemma-4-26b-a4b-it/) | 73.2% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemma 4 26B A4B |
| [Qwen3-Max](https://meirinho.com.br/en/ia/modelos/qwen3-max/) | 72.6% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3-Max |
| [Claude Haiku 4.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-4-5/) | 71.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude Haiku 4.5 |
| [GPT-5 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-nano/) | 69.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-5 Nano |
| [GPT-4.5](https://meirinho.com.br/en/ia/modelos/gpt-4-5/) | 68.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-4.5 |
| [DeepSeek V3](https://meirinho.com.br/en/ia/modelos/deepseek-v3/) | 67.6% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek V3 |
| [Llama 4 Maverick](https://meirinho.com.br/en/ia/modelos/llama-4-maverick/) | 67% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 4 Maverick |
| [GPT-4.1](https://meirinho.com.br/en/ia/modelos/gpt-4-1/) | 66.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-4.1 |
| [GPT-4.1 Mini](https://meirinho.com.br/en/ia/modelos/gpt-4-1-mini/) | 65.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-4.1 Mini |
| [Qwen3 32B](https://meirinho.com.br/en/ia/modelos/qwen3-32b/) | 65.7% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3 32B |
| [Gemini 2.0 Pro](https://meirinho.com.br/en/ia/modelos/gemini-2-0-pro/) | 65.7% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 2.0 Pro |
| [QWQ-Plus](https://meirinho.com.br/en/ia/modelos/qwq-plus/) | 65.4% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | QWQ-Plus |
| [QwQ-32B](https://meirinho.com.br/en/ia/modelos/qwq-32b/) | 65.3% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | QwQ-32B |
| [DeepSeek-R1-Distill-Qwen-32B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-qwen-32b/) | 64.1% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek-R1-Distill-Qwen-32B |
| [Gemini 2.0 Flash](https://meirinho.com.br/en/ia/modelos/gemini-2-0-flash/) | 64.1% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 2.0 Flash |
| [Qwen3 14B](https://meirinho.com.br/en/ia/modelos/qwen3-14b/) | 63.8% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3 14B |
| [o1-mini](https://meirinho.com.br/en/ia/modelos/o1-mini/) | 62.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | o1-mini |
| [Qwen3 30B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-30b-a3b/) | 61.7% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3 30B A3B |
| [gpt-oss-20b](https://meirinho.com.br/en/ia/modelos/gpt-oss-20b/) | 60.8% Epoch AI | OpenAI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | gpt-oss-20b |
| [Mistral Medium 3](https://meirinho.com.br/en/ia/modelos/mistral-medium-3/) | 59.5% Epoch AI | Mistral AI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Medium 3 |
| [Gemini 1.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-1-5-pro/) | 57.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 1.5 Pro |
| [Gemini 2.0 Flash Thinking](https://meirinho.com.br/en/ia/modelos/gemini-2-0-flash-thinking/) | 57.1% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 2.0 Flash Thinking |
| [Qwen3-8B](https://meirinho.com.br/en/ia/modelos/qwen3-8b/) | 56.8% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3-8B |
| [Phi 4](https://meirinho.com.br/en/ia/modelos/phi-4/) | 56.1% Epoch AI | Microsoft | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Phi 4 |
| [Magistral Small 1.0](https://meirinho.com.br/en/ia/modelos/magistral-small-1-0/) | 56.1% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Magistral Small 1.0 |
| [Qwen2.5-Max](https://meirinho.com.br/en/ia/modelos/qwen2-5-max/) | 56.1% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen2.5-Max |
| [DeepSeek-R1-Distill-Llama-70B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-llama-70b/) | 55.7% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek-R1-Distill-Llama-70B |
| [Claude 3.5 Sonnet (October 2024)](https://meirinho.com.br/en/ia/modelos/claude-3-5-sonnet-october-2024/) | 55.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 3.5 Sonnet (October 2024) |
| [Claude 3.5 Sonnet](https://meirinho.com.br/en/ia/modelos/claude-3-5-sonnet/) | 54% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 3.5 Sonnet |
| [Grok-2](https://meirinho.com.br/en/ia/modelos/grok-2/) | 53.8% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Grok-2 |
| [Qwen3-4B](https://meirinho.com.br/en/ia/modelos/qwen3-4b/) | 52.3% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3-4B |
| [Llama 4 Scout](https://meirinho.com.br/en/ia/modelos/llama-4-scout/) | 51.8% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 4 Scout |
| [Mistral Large 2](https://meirinho.com.br/en/ia/modelos/mistral-large-2/) | 51.3% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Large 2 |
| [Llama 3.1-405B](https://meirinho.com.br/en/ia/modelos/llama-3-1-405b/) | 50.9% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3.1-405B |
| [o1-preview](https://meirinho.com.br/en/ia/modelos/o1-preview/) | 50.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | o1-preview |
| [Mistral Small 3.2](https://meirinho.com.br/en/ia/modelos/mistral-small-3-2/) | 49.1% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Small 3.2 |
| [Qwen2.5-72B](https://meirinho.com.br/en/ia/modelos/qwen2-5-72b/) | 49.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen2.5-72B |
| [GPT-4.1 Nano](https://meirinho.com.br/en/ia/modelos/gpt-4-1-nano/) | 48.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-4.1 Nano |
| [Qwen-Plus](https://meirinho.com.br/en/ia/modelos/qwen-plus/) | 48.1% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen-Plus |
| [Gemma 3 27B](https://meirinho.com.br/en/ia/modelos/gemma-3-27b-it/) | 47.7% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemma 3 27B |
| [Magistral Small 1.2](https://meirinho.com.br/en/ia/modelos/magistral-small-1-2/) | 47.6% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Magistral Small 1.2 |
| [Mistral Small 3.1](https://meirinho.com.br/en/ia/modelos/mistral-small-3-1/) | 47.5% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Small 3.1 |
| [Llama 3.3 70B](https://meirinho.com.br/en/ia/modelos/llama-3-3-70b/) | 47.4% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3.3 70B |
| [Mistral Small 3](https://meirinho.com.br/en/ia/modelos/mistral-small-24b-instruct-2501/) | 47.3% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Small 3 |
| [Gemini 1.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-1-5-flash/) | 47.3% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 1.5 Flash |
| [Claude 3 Opus](https://meirinho.com.br/en/ia/modelos/claude-3-opus/) | 47.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 3 Opus |
| [GPT-4 Turbo (Apr 2024)](https://meirinho.com.br/en/ia/modelos/gpt-4-turbo-apr-2024/) | 46.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-4 Turbo (Apr 2024) |
| [Tulu 3 (Tülu 3) 70B](https://meirinho.com.br/en/ia/modelos/tulu-3-t-lu-3-70b/) | 46.3% Epoch AI | Ai2 (Allen Institute) | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Tulu 3 (Tülu 3) 70B |
| [Qwen2.5-32B](https://meirinho.com.br/en/ia/modelos/qwen2-5-32b/) | 46.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen2.5-32B |
| [DeepSeek-R1-Distill-Qwen-14B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-qwen-14b/) | 44.7% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek-R1-Distill-Qwen-14B |
| [Llama 3.1-70B](https://meirinho.com.br/en/ia/modelos/llama-3-1-70b/) | 44.2% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3.1-70B |
| [WizardLM-2 8x22B](https://meirinho.com.br/en/ia/modelos/wizardlm-2-8x22b/) | 43.4% Epoch AI | Microsoft | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | WizardLM-2 8x22B |
| [GPT-4 Turbo (Nov 2023)](https://meirinho.com.br/en/ia/modelos/gpt-4-turbo-nov-2023/) | 42.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-4 Turbo (Nov 2023) |
| [Qwen-Turbo](https://meirinho.com.br/en/ia/modelos/qwen-turbo/) | 41.8% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen-Turbo |
| [Llama 3.2 90B](https://meirinho.com.br/en/ia/modelos/llama-3-2-90b/) | 41% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3.2 90B |
| [Qwen2-72B](https://meirinho.com.br/en/ia/modelos/qwen2-72b/) | 40.8% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen2-72B |
| [Claude 3 Sonnet](https://meirinho.com.br/en/ia/modelos/claude-3-sonnet/) | 40.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 3 Sonnet |
| [Llama 3-70B](https://meirinho.com.br/en/ia/modelos/llama-3-70b/) | 40.6% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3-70B |
| [Gemma 3 12B](https://meirinho.com.br/en/ia/modelos/gemma-3-12b-it/) | 39.5% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemma 3 12B |
| [Mistral Large](https://meirinho.com.br/en/ia/modelos/mistral-large/) | 38.8% Epoch AI | Mistral AI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Large |
| [Claude 3.5 Haiku](https://meirinho.com.br/en/ia/modelos/claude-3-5-haiku/) | 38.1% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 3.5 Haiku |
| [Qwen3-1.7B](https://meirinho.com.br/en/ia/modelos/qwen3-1-7b/) | 38% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen3-1.7B |
| [GPT-4o-mini](https://meirinho.com.br/en/ia/modelos/gpt-4o-mini/) | 37.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | GPT-4o-mini |
| [Gemma 2 27B](https://meirinho.com.br/en/ia/modelos/gemma-2-27b-it/) | 36.5% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemma 2 27B |
| [Claude 3 Haiku](https://meirinho.com.br/en/ia/modelos/claude-3-haiku/) | 36.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 3 Haiku |
| [Qwen2.5-7B](https://meirinho.com.br/en/ia/modelos/qwen2-5-7b/) | 35.5% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen2.5-7B |
| [Claude 2](https://meirinho.com.br/en/ia/modelos/claude-2/) | 34.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 2 |
| [Mixtral 8x22B](https://meirinho.com.br/en/ia/modelos/mixtral-8x22b/) | 34.1% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mixtral 8x22B |
| [Gemini 1.0 Pro](https://meirinho.com.br/en/ia/modelos/gemini-1-0-pro/) | 34% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 1.0 Pro |
| [DeepSeek-R1-Distill-Qwen-1.5B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-qwen-1-5b/) | 33.6% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek-R1-Distill-Qwen-1.5B |
| [Claude 2.1](https://meirinho.com.br/en/ia/modelos/claude-2-1/) | 33% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Claude 2.1 |
| [Gemini 1.5 Flash 8B](https://meirinho.com.br/en/ia/modelos/gemini-1-5-flash-8b/) | 33% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemini 1.5 Flash 8B |
| [Yi-1.5-34B](https://meirinho.com.br/en/ia/modelos/yi-1-5-34b/) | 32% Epoch AI | 01.AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Yi-1.5-34B |
| [Qwen1.5-32B](https://meirinho.com.br/en/ia/modelos/qwen1-5-32b/) | 30.7% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen1.5-32B |
| [Mixtral 8x7B](https://meirinho.com.br/en/ia/modelos/mixtral-8x7b/) | 30.6% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mixtral 8x7B |
| [Mistral Nemo](https://meirinho.com.br/en/ia/modelos/mistral-nemo/) | 29.9% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral Nemo |
| [Qwen1.5-72B](https://meirinho.com.br/en/ia/modelos/qwen1-5-72b/) | 28.8% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Qwen1.5-72B |
| [phi-3-medium 14B](https://meirinho.com.br/en/ia/modelos/phi-3-medium-14b/) | 27.6% Epoch AI | Microsoft | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | phi-3-medium 14B |
| [Gemma 2 9B](https://meirinho.com.br/en/ia/modelos/gemma-2-9b/) | 27.5% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemma 2 9B |
| [Ministral 8B](https://meirinho.com.br/en/ia/modelos/ministral-8b/) | 27.1% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Ministral 8B |
| [Llama 3.1-8B](https://meirinho.com.br/en/ia/modelos/llama-3-1-8b/) | 27% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3.1-8B |
| [Llama 2-70B](https://meirinho.com.br/en/ia/modelos/llama-2-70b/) | 26.3% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 2-70B |
| [Llama 3-8B](https://meirinho.com.br/en/ia/modelos/llama-3-8b/) | 26.1% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3-8B |
| [Ministral 3B](https://meirinho.com.br/en/ia/modelos/ministral-3b/) | 25.3% Epoch AI | Mistral AI | closed | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Ministral 3B |
| [DeepSeek LLM 67B](https://meirinho.com.br/en/ia/modelos/deepseek-llm-67b/) | 24.6% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | DeepSeek LLM 67B |
| [Llama 3.2 1B](https://meirinho.com.br/en/ia/modelos/llama-3-2-1b/) | 23.9% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Llama 3.2 1B |
| [Gemma 3 4B](https://meirinho.com.br/en/ia/modelos/gemma-3-4b-it/) | 23.2% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemma 3 4B |
| [Gemma 3 1B](https://meirinho.com.br/en/ia/modelos/gemma-3-1b/) | 19.9% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Gemma 3 1B |
| [Mistral 7B](https://meirinho.com.br/en/ia/modelos/mistral-7b/) | 15.2% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Mistral 7B |
| [Yi-34B](https://meirinho.com.br/en/ia/modelos/yi-34b/) | 14.7% Epoch AI | 01.AI | open | [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond) | Yi-34B |

## [OTIS Mock AIME](https://meirinho.com.br/en/ia/modelos/benchmark/otis-aime/)

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/texto/)

| Model | % | Lab | Weights | Source | Compare models |
| --- | --- | --- | --- | --- | --- |
| [Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/claude-opus-5-5/) | 100% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 5.5 |
| [Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5-5/) | 100% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Sonnet 5.5 |
| [GPT-6.1 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-1-sol/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-6.1 Sol |
| [GPT-6 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-astra/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-6 Astra |
| [GPT-5.5](https://meirinho.com.br/en/ia/modelos/gpt-5-5/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.5 |
| [Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/gemini-4-argon/) | 100% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 4 Argon |
| [Qwen3.8 Max](https://meirinho.com.br/en/ia/modelos/qwen3-8-max/) | 100% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.8 Max |
| [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) | 100% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Kimi K3.1 Pro |
| [Claude Fable 5](https://meirinho.com.br/en/ia/modelos/claude-fable-5/) | 100% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Fable 5 |
| [Claude Fable 5.1](https://meirinho.com.br/en/ia/modelos/claude-fable-5-1/) | 100% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Fable 5.1 |
| [GPT-5.5 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-5-pro/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.5 Pro |
| [GPT-5.6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.6 Sol |
| [GPT-6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-sol/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-6 Sol |
| [GPT-6.1 Terra](https://meirinho.com.br/en/ia/modelos/gpt-6-1-terra/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-6.1 Terra |
| [Gemini 4 Carbon](https://meirinho.com.br/en/ia/modelos/gemini-4-carbon/) | 100% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 4 Carbon |
| [GPT-6.2 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-2-astra/) | 100% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-6.2 Astra |
| [Claude Fable 5.2](https://meirinho.com.br/en/ia/modelos/claude-fable-5-2/) | 99.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Fable 5.2 |
| [GPT-5.6 Terra](https://meirinho.com.br/en/ia/modelos/gpt-5-6-terra/) | 99.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.6 Terra |
| [Mistral Large 4](https://meirinho.com.br/en/ia/modelos/mistral-large-4/) | 99.4% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral Large 4 |
| [Muse Wonder](https://meirinho.com.br/en/ia/modelos/muse-wonder/) | 99.4% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Muse Wonder |
| [Grok 4.6](https://meirinho.com.br/en/ia/modelos/grok-4-6/) | 99.2% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok 4.6 |
| [Muse Spark 1.3](https://meirinho.com.br/en/ia/modelos/muse-spark-1-3/) | 99.2% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Muse Spark 1.3 |
| [Doubao Seed Evolving (Nov 2026)](https://meirinho.com.br/en/ia/modelos/doubao-seed-evolving-nov-2026/) | 99.2% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Doubao Seed Evolving (Nov 2026) |
| [Claude Haiku 5.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-5-5/) | 98.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Haiku 5.5 |
| [GPT-6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-6-luna/) | 98.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-6 Luna |
| [Gemini 3.8 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-8-flash/) | 98.9% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3.8 Flash |
| [Claude Opus 5](https://meirinho.com.br/en/ia/modelos/claude-opus-5/) | 98.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 5 |
| [Claude Opus 4.8](https://meirinho.com.br/en/ia/modelos/claude-opus-4-8/) | 98.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 4.8 |
| [DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/deepseek-v4-1-flash/) | 98.3% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek V4.1 Flash |
| [GPT-5.6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna/) | 98.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.6 Luna |
| [Grok 4.7](https://meirinho.com.br/en/ia/modelos/grok-4-7/) | 98.1% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok 4.7 |
| [Muse Spark Open](https://meirinho.com.br/en/ia/modelos/muse-spark-open/) | 98.1% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Muse Spark Open |
| [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) | 97.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 4.7 |
| [GPT-5.4](https://meirinho.com.br/en/ia/modelos/gpt-5-4/) | 97.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.4 |
| [Grok 4.5](https://meirinho.com.br/en/ia/modelos/grok-4-5/) | 97.8% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok 4.5 |
| [Kimi K3](https://meirinho.com.br/en/ia/modelos/kimi-k3/) | 97.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Kimi K3 |
| [Gemini 3.7 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-7-flash/) | 97.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3.7 Flash |
| [DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/deepseek-v4-pro/) | 96.7% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek V4 Pro |
| [GPT-5.2](https://meirinho.com.br/en/ia/modelos/gpt-5-2/) | 96.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.2 |
| [Kimi K2.6](https://meirinho.com.br/en/ia/modelos/kimi-k2-6/) | 96.1% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Kimi K2.6 |
| [Doubao Seed 2.1 Turbo](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-1-turbo/) | 96.1% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Doubao Seed 2.1 Turbo |
| [Gemini 3.1 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-1-pro/) | 95.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3.1 Pro |
| [Gemini 3.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash/) | 95.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3.5 Flash |
| [Kimi K2.7 Code](https://meirinho.com.br/en/ia/modelos/kimi-k2-7-code/) | 95.6% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Kimi K2.7 Code |
| [Qwen3.7 Max](https://meirinho.com.br/en/ia/modelos/qwen3-7-max/) | 95.6% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.7 Max |
| [Doubao Seed 2.1 Pro](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-1-pro/) | 95.6% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Doubao Seed 2.1 Pro |
| [Gemini 3 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-flash/) | 95.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3 Flash |
| [Claude Sonnet 5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5/) | 94.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Sonnet 5 |
| [Claude Opus 4.6](https://meirinho.com.br/en/ia/modelos/claude-opus-4-6/) | 94.4% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 4.6 |
| [DeepSeek V4 Flash 0731](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash-0731/) | 94.4% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek V4 Flash 0731 |
| [Gemini 3.6 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-6-flash/) | 94.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3.6 Flash |
| [GLM-5.3-Flash](https://meirinho.com.br/en/ia/modelos/glm-5-3-flash/) | 93.9% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GLM-5.3-Flash |
| [Doubao-Seed-2.0 Pro](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-0-pro/) | 93.9% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Doubao-Seed-2.0 Pro |
| [Qwen3.6 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-6-plus/) | 93.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.6 Plus |
| [Qwen3.7 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-7-plus/) | 93.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.7 Plus |
| [GLM 5.1](https://meirinho.com.br/en/ia/modelos/glm-5-1/) | 93.3% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GLM 5.1 |
| [Grok 4.3 Beta](https://meirinho.com.br/en/ia/modelos/grok-4-3-beta/) | 93.3% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok 4.3 Beta |
| [Kimi K2.5](https://meirinho.com.br/en/ia/modelos/kimi-k2-5/) | 92.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Kimi K2.5 |
| [Grok 4.20](https://meirinho.com.br/en/ia/modelos/grok-4-20/) | 92.2% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok 4.20 |
| [GPT-5](https://meirinho.com.br/en/ia/modelos/gpt-5/) | 91.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5 |
| [Gemini 3 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-pro/) | 91.4% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3 Pro |
| [GLM-5.3](https://meirinho.com.br/en/ia/modelos/glm-5-3/) | 91.1% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GLM-5.3 |
| [Qwen3.6 27B](https://meirinho.com.br/en/ia/modelos/qwen3-6-27b/) | 91.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.6 27B |
| [Doubao Seed 2.0 Lite (Apr 2026)](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-0-lite-apr-2026/) | 91.1% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Doubao Seed 2.0 Lite (Apr 2026) |
| [Qwen 3.6 Max (Preview)](https://meirinho.com.br/en/ia/modelos/qwen-3-6-max-preview/) | 91.1% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen 3.6 Max (Preview) |
| [gpt-oss-120b](https://meirinho.com.br/en/ia/modelos/gpt-oss-120b/) | 88.9% Epoch AI | OpenAI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | gpt-oss-120b |
| [GPT-5.4 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-4-mini/) | 88.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.4 Mini |
| [Qwen3.5 397B A17B](https://meirinho.com.br/en/ia/modelos/qwen3-5-397b-a17b/) | 88.9% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.5 397B A17B |
| [Muse Spark](https://meirinho.com.br/en/ia/modelos/muse-spark/) | 88.9% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Muse Spark |
| [GPT-5.1](https://meirinho.com.br/en/ia/modelos/gpt-5-1/) | 88.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.1 |
| [DeepSeek V3.2](https://meirinho.com.br/en/ia/modelos/deepseek-v3-2/) | 87.8% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek V3.2 |
| [GPT-5.4 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-4-nano/) | 87.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.4 Nano |
| [Qwen3-235B-A22B](https://meirinho.com.br/en/ia/modelos/qwen3-235b-a22b/) | 86.7% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3-235B-A22B |
| [Nemotron 3 Ultra](https://meirinho.com.br/en/ia/modelos/nemotron-3-ultra/) | 86.7% Epoch AI | NVIDIA | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Nemotron 3 Ultra |
| [GPT-5 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-mini/) | 86.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5 Mini |
| [Qwen3.5 Plus 2026-02-15](https://meirinho.com.br/en/ia/modelos/qwen3-5-plus-02-15/) | 86.7% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.5 Plus 2026-02-15 |
| [Qwen3.6 35B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-6-35b-a3b/) | 86.7% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.6 35B A3B |
| [Qwen3.7 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-7-flash/) | 86.7% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.7 Flash |
| [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) | 86.4% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GLM-5.2 |
| [Claude Opus 4.5](https://meirinho.com.br/en/ia/modelos/claude-opus-4-5/) | 86.1% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 4.5 |
| [Claude Sonnet 4.6](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-6/) | 85.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Sonnet 4.6 |
| [Gemini 2.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-2-5-pro/) | 84.7% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 2.5 Pro |
| [o3](https://meirinho.com.br/en/ia/modelos/o3/) | 84.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | o3 |
| [Qwen3.5-Flash](https://meirinho.com.br/en/ia/modelos/qwen3-5-flash-02-23/) | 84.4% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.5-Flash |
| [Qwen3.6 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-6-flash/) | 84.4% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.6 Flash |
| [Grok 4](https://meirinho.com.br/en/ia/modelos/grok-4/) | 84% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok 4 |
| [GLM 4.7](https://meirinho.com.br/en/ia/modelos/glm-4-7/) | 83.3% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GLM 4.7 |
| [Kimi K2 Thinking](https://meirinho.com.br/en/ia/modelos/kimi-k2-thinking/) | 83.1% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Kimi K2 Thinking |
| [Gemma 4 26B A4B](https://meirinho.com.br/en/ia/modelos/gemma-4-26b-a4b-it/) | 82.2% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemma 4 26B A4B |
| [o4-mini](https://meirinho.com.br/en/ia/modelos/o4-mini/) | 81.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | o4-mini |
| [GPT-5 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-nano/) | 81.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5 Nano |
| [Gemini 3.1 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-1-flash-lite/) | 80% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3.1 Flash Lite |
| [GLM 5](https://meirinho.com.br/en/ia/modelos/glm-5/) | 80% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GLM 5 |
| [Claude Sonnet 4.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-5/) | 77.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Sonnet 4.5 |
| [Grok-3 mini](https://meirinho.com.br/en/ia/modelos/grok-3-mini/) | 77.8% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok-3 mini |
| [o3-mini](https://meirinho.com.br/en/ia/modelos/o3-mini/) | 76.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | o3-mini |
| [Qwen3-Max](https://meirinho.com.br/en/ia/modelos/qwen3-max/) | 73.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3-Max |
| [o1](https://meirinho.com.br/en/ia/modelos/o1/) | 73.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | o1 |
| [MiniMax M3](https://meirinho.com.br/en/ia/modelos/minimax-m3/) | 71.1% Epoch AI | MiniMax | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | MiniMax M3 |
| [Gemini 3.5 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash-lite/) | 71.1% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 3.5 Flash Lite |
| [Claude Sonnet 4](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4/) | 71.1% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Sonnet 4 |
| [Qwen3.5-35B-A3B](https://meirinho.com.br/en/ia/modelos/qwen3-5-35b-a3b/) | 70% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.5-35B-A3B |
| [Doubao-Seed-1.6](https://meirinho.com.br/en/ia/modelos/doubao-seed-1-6/) | 69.4% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Doubao-Seed-1.6 |
| [Claude Opus 4.1](https://meirinho.com.br/en/ia/modelos/claude-opus-4-1/) | 68.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 4.1 |
| [GPT-5.5 Instant](https://meirinho.com.br/en/ia/modelos/gpt-5-5-instant/) | 68.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-5.5 Instant |
| [Qwen3 32B](https://meirinho.com.br/en/ia/modelos/qwen3-32b/) | 66.9% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3 32B |
| [Claude Haiku 4.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-4-5/) | 66.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Haiku 4.5 |
| [DeepSeek R1](https://meirinho.com.br/en/ia/modelos/deepseek-r1/) | 66.4% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek R1 |
| [Qwen3 14B](https://meirinho.com.br/en/ia/modelos/qwen3-14b/) | 66.4% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3 14B |
| [gpt-oss-20b](https://meirinho.com.br/en/ia/modelos/gpt-oss-20b/) | 65.3% Epoch AI | OpenAI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | gpt-oss-20b |
| [Claude Opus 4](https://meirinho.com.br/en/ia/modelos/claude-opus-4/) | 64.4% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude Opus 4 |
| [Qwen3 30B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-30b-a3b/) | 62.8% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3 30B A3B |
| [Qwen3.5-9B](https://meirinho.com.br/en/ia/modelos/qwen3-5-9b/) | 61.7% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.5-9B |
| [QwQ-32B](https://meirinho.com.br/en/ia/modelos/qwq-32b/) | 59.2% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | QwQ-32B |
| [Claude 3.7 Sonnet](https://meirinho.com.br/en/ia/modelos/claude-3-7-sonnet/) | 57.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 3.7 Sonnet |
| [Gemini 2.0 Flash Thinking](https://meirinho.com.br/en/ia/modelos/gemini-2-0-flash-thinking/) | 57.8% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 2.0 Flash Thinking |
| [Qwen3-8B](https://meirinho.com.br/en/ia/modelos/qwen3-8b/) | 56.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3-8B |
| [Qwen3.5-4B](https://meirinho.com.br/en/ia/modelos/qwen3-5-4b/) | 55.8% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3.5-4B |
| [DeepSeek-R1-Distill-Qwen-32B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-qwen-32b/) | 55.6% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek-R1-Distill-Qwen-32B |
| [Grok 3](https://meirinho.com.br/en/ia/modelos/grok-3/) | 55.6% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok 3 |
| [DeepSeek-R1-Distill-Llama-70B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-llama-70b/) | 51.4% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek-R1-Distill-Llama-70B |
| [DeepSeek-R1-Distill-Qwen-14B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-qwen-14b/) | 50.6% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek-R1-Distill-Qwen-14B |
| [o1-mini](https://meirinho.com.br/en/ia/modelos/o1-mini/) | 46.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | o1-mini |
| [GPT-4.1 Mini](https://meirinho.com.br/en/ia/modelos/gpt-4-1-mini/) | 44.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-4.1 Mini |
| [GPT-4.1](https://meirinho.com.br/en/ia/modelos/gpt-4-1/) | 38.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-4.1 |
| [DeepSeek V3](https://meirinho.com.br/en/ia/modelos/deepseek-v3/) | 37.8% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek V3 |
| [GPT-4.5](https://meirinho.com.br/en/ia/modelos/gpt-4-5/) | 37.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-4.5 |
| [Mistral Medium 3](https://meirinho.com.br/en/ia/modelos/mistral-medium-3/) | 32.2% Epoch AI | Mistral AI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral Medium 3 |
| [Gemini 2.0 Flash](https://meirinho.com.br/en/ia/modelos/gemini-2-0-flash/) | 31.1% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 2.0 Flash |
| [o1-preview](https://meirinho.com.br/en/ia/modelos/o1-preview/) | 31.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | o1-preview |
| [Mistral Small 3.2](https://meirinho.com.br/en/ia/modelos/mistral-small-3-2/) | 30.3% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral Small 3.2 |
| [Magistral Small 1.0](https://meirinho.com.br/en/ia/modelos/magistral-small-1-0/) | 30% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Magistral Small 1.0 |
| [GPT-4.1 Nano](https://meirinho.com.br/en/ia/modelos/gpt-4-1-nano/) | 28.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-4.1 Nano |
| [Magistral Small 1.2](https://meirinho.com.br/en/ia/modelos/magistral-small-1-2/) | 28.1% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Magistral Small 1.2 |
| [Gemini 1.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-1-5-pro/) | 23.1% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 1.5 Pro |
| [Gemma 3 27B](https://meirinho.com.br/en/ia/modelos/gemma-3-27b-it/) | 22.5% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemma 3 27B |
| [DeepSeek-R1-Distill-Qwen-1.5B](https://meirinho.com.br/en/ia/modelos/deepseek-r1-distill-qwen-1-5b/) | 21.4% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek-R1-Distill-Qwen-1.5B |
| [Llama 4 Maverick](https://meirinho.com.br/en/ia/modelos/llama-4-maverick/) | 20.6% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 4 Maverick |
| [Qwen-Plus](https://meirinho.com.br/en/ia/modelos/qwen-plus/) | 17.8% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen-Plus |
| [Gemma 3 12B](https://meirinho.com.br/en/ia/modelos/gemma-3-12b-it/) | 16.7% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemma 3 12B |
| [Gemini 1.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-1-5-flash/) | 16.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 1.5 Flash |
| [Qwen2.5-Max](https://meirinho.com.br/en/ia/modelos/qwen2-5-max/) | 16.1% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen2.5-Max |
| [Phi 4](https://meirinho.com.br/en/ia/modelos/phi-4/) | 13.8% Epoch AI | Microsoft | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Phi 4 |
| [Grok-2](https://meirinho.com.br/en/ia/modelos/grok-2/) | 11.5% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Grok-2 |
| [Llama 3.1-405B](https://meirinho.com.br/en/ia/modelos/llama-3-1-405b/) | 9.7% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3.1-405B |
| [Claude 3.5 Sonnet (October 2024)](https://meirinho.com.br/en/ia/modelos/claude-3-5-sonnet-october-2024/) | 8.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 3.5 Sonnet (October 2024) |
| [Qwen2.5-72B](https://meirinho.com.br/en/ia/modelos/qwen2-5-72b/) | 8.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen2.5-72B |
| [Qwen3-1.7B](https://meirinho.com.br/en/ia/modelos/qwen3-1-7b/) | 8.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen3-1.7B |
| [Llama 4 Scout](https://meirinho.com.br/en/ia/modelos/llama-4-scout/) | 7.8% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 4 Scout |
| [Mistral Large 2](https://meirinho.com.br/en/ia/modelos/mistral-large-2/) | 7.8% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral Large 2 |
| [Gemma 3 4B](https://meirinho.com.br/en/ia/modelos/gemma-3-4b-it/) | 7.5% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemma 3 4B |
| [Qwen2.5-32B](https://meirinho.com.br/en/ia/modelos/qwen2-5-32b/) | 7.4% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen2.5-32B |
| [GPT-4o-mini](https://meirinho.com.br/en/ia/modelos/gpt-4o-mini/) | 6.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-4o-mini |
| [Mistral Small 3](https://meirinho.com.br/en/ia/modelos/mistral-small-24b-instruct-2501/) | 6.7% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral Small 3 |
| [GPT-4 Turbo (Apr 2024)](https://meirinho.com.br/en/ia/modelos/gpt-4-turbo-apr-2024/) | 6.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | GPT-4 Turbo (Apr 2024) |
| [Claude 3.5 Sonnet](https://meirinho.com.br/en/ia/modelos/claude-3-5-sonnet/) | 6.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 3.5 Sonnet |
| [Qwen-Turbo](https://meirinho.com.br/en/ia/modelos/qwen-turbo/) | 6.1% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen-Turbo |
| [Mistral Small 3.1](https://meirinho.com.br/en/ia/modelos/mistral-small-3-1/) | 5.8% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral Small 3.1 |
| [Llama 3.3 70B](https://meirinho.com.br/en/ia/modelos/llama-3-3-70b/) | 5.1% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3.3 70B |
| [Claude 3 Opus](https://meirinho.com.br/en/ia/modelos/claude-3-opus/) | 4.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 3 Opus |
| [Gemini 1.5 Flash 8B](https://meirinho.com.br/en/ia/modelos/gemini-1-5-flash-8b/) | 4.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 1.5 Flash 8B |
| [Tulu 3 (Tülu 3) 70B](https://meirinho.com.br/en/ia/modelos/tulu-3-t-lu-3-70b/) | 4.4% Epoch AI | Ai2 (Allen Institute) | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Tulu 3 (Tülu 3) 70B |
| [Claude 3.5 Haiku](https://meirinho.com.br/en/ia/modelos/claude-3-5-haiku/) | 4.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 3.5 Haiku |
| [Llama 3-70B](https://meirinho.com.br/en/ia/modelos/llama-3-70b/) | 4.3% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3-70B |
| [Llama 3.1-70B](https://meirinho.com.br/en/ia/modelos/llama-3-1-70b/) | 3.6% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3.1-70B |
| [Llama 3.2 90B](https://meirinho.com.br/en/ia/modelos/llama-3-2-90b/) | 2.6% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3.2 90B |
| [Claude 2](https://meirinho.com.br/en/ia/modelos/claude-2/) | 2.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 2 |
| [Claude 3 Sonnet](https://meirinho.com.br/en/ia/modelos/claude-3-sonnet/) | 2.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 3 Sonnet |
| [Qwen2.5-7B](https://meirinho.com.br/en/ia/modelos/qwen2-5-7b/) | 2.5% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Qwen2.5-7B |
| [Mistral Large](https://meirinho.com.br/en/ia/modelos/mistral-large/) | 1.9% Epoch AI | Mistral AI | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral Large |
| [Claude 2.1](https://meirinho.com.br/en/ia/modelos/claude-2-1/) | 1.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 2.1 |
| [Llama 3-8B](https://meirinho.com.br/en/ia/modelos/llama-3-8b/) | 1.9% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3-8B |
| [Claude 3 Haiku](https://meirinho.com.br/en/ia/modelos/claude-3-haiku/) | 1.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Claude 3 Haiku |
| [Llama 3.1-8B](https://meirinho.com.br/en/ia/modelos/llama-3-1-8b/) | 1.7% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3.1-8B |
| [Gemma 2 27B](https://meirinho.com.br/en/ia/modelos/gemma-2-27b-it/) | 1.4% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemma 2 27B |
| [Gemini 1.0 Pro](https://meirinho.com.br/en/ia/modelos/gemini-1-0-pro/) | 1.1% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemini 1.0 Pro |
| [Gemma 3 1B](https://meirinho.com.br/en/ia/modelos/gemma-3-1b/) | 1.1% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemma 3 1B |
| [DeepSeek LLM 67B](https://meirinho.com.br/en/ia/modelos/deepseek-llm-67b/) | 0.8% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | DeepSeek LLM 67B |
| [Gemma 2 9B](https://meirinho.com.br/en/ia/modelos/gemma-2-9b/) | 0.6% Epoch AI | Google | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Gemma 2 9B |
| [Llama 3.2 1B](https://meirinho.com.br/en/ia/modelos/llama-3-2-1b/) | 0.6% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 3.2 1B |
| [Mistral 7B](https://meirinho.com.br/en/ia/modelos/mistral-7b/) | 0.3% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Mistral 7B |
| [Llama 2-70B](https://meirinho.com.br/en/ia/modelos/llama-2-70b/) | 0% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025) | Llama 2-70B |

## [FrontierMath (tiers 1 to 3)](https://meirinho.com.br/en/ia/modelos/benchmark/frontiermath/)

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/texto/)

| Model | % | Lab | Weights | Source | Compare models |
| --- | --- | --- | --- | --- | --- |
| [Claude Fable 5.2](https://meirinho.com.br/en/ia/modelos/claude-fable-5-2/) | 95.1% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Fable 5.2 |
| [GPT-6.1 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-1-sol/) | 93.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-6.1 Sol |
| [GPT-6 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-astra/) | 93.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-6 Astra |
| [GPT-6.2 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-2-astra/) | 92.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-6.2 Astra |
| [Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/claude-opus-5-5/) | 91.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Opus 5.5 |
| [GPT-6.1 Terra](https://meirinho.com.br/en/ia/modelos/gpt-6-1-terra/) | 90.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-6.1 Terra |
| [Gemini 4 Carbon](https://meirinho.com.br/en/ia/modelos/gemini-4-carbon/) | 90.5% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 4 Carbon |
| [Claude Fable 5.1](https://meirinho.com.br/en/ia/modelos/claude-fable-5-1/) | 90.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Fable 5.1 |
| [GPT-6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-sol/) | 89.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-6 Sol |
| [GPT-5.6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol/) | 89.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.6 Sol |
| [Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5-5/) | 88.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Sonnet 5.5 |
| [GPT-5.5 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-5-pro/) | 87.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.5 Pro |
| [Claude Fable 5](https://meirinho.com.br/en/ia/modelos/claude-fable-5/) | 87% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Fable 5 |
| [Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/gemini-4-argon/) | 86% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 4 Argon |
| [GPT-5.6 Terra](https://meirinho.com.br/en/ia/modelos/gpt-5-6-terra/) | 86% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.6 Terra |
| [Claude Opus 5](https://meirinho.com.br/en/ia/modelos/claude-opus-5/) | 85.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Opus 5 |
| [GPT-5.5](https://meirinho.com.br/en/ia/modelos/gpt-5-5/) | 85.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.5 |
| [GPT-5.4 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-4-pro/) | 82.5% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.4 Pro |
| [Muse Wonder](https://meirinho.com.br/en/ia/modelos/muse-wonder/) | 82.1% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Muse Wonder |
| [GPT-5.6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna/) | 82.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.6 Luna |
| [Claude Opus 4.8](https://meirinho.com.br/en/ia/modelos/claude-opus-4-8/) | 80% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Opus 4.8 |
| [GPT-6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-6-luna/) | 78.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-6 Luna |
| [GPT-5.4](https://meirinho.com.br/en/ia/modelos/gpt-5-4/) | 78.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.4 |
| [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) | 78.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Kimi K3.1 Pro |
| [Claude Haiku 5.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-5-5/) | 75.1% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Haiku 5.5 |
| [Muse Spark 1.3](https://meirinho.com.br/en/ia/modelos/muse-spark-1-3/) | 74.4% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Muse Spark 1.3 |
| [GPT-5.2 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-2-pro/) | 74% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.2 Pro |
| [Kimi K3](https://meirinho.com.br/en/ia/modelos/kimi-k3/) | 72.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Kimi K3 |
| [Gemini 3.7 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-7-flash/) | 71.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3.7 Flash |
| [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) | 70.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Opus 4.7 |
| [Muse Spark Open](https://meirinho.com.br/en/ia/modelos/muse-spark-open/) | 69.8% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Muse Spark Open |
| [GLM-5.3](https://meirinho.com.br/en/ia/modelos/glm-5-3/) | 68.8% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GLM-5.3 |
| [Gemini 3.8 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-8-flash/) | 68.4% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3.8 Flash |
| [GPT-5.2](https://meirinho.com.br/en/ia/modelos/gpt-5-2/) | 67.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.2 |
| [DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/deepseek-v4-1-flash/) | 67.4% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | DeepSeek V4.1 Flash |
| [Grok 4.6](https://meirinho.com.br/en/ia/modelos/grok-4-6/) | 66% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Grok 4.6 |
| [Claude Opus 4.6](https://meirinho.com.br/en/ia/modelos/claude-opus-4-6/) | 66% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Opus 4.6 |
| [Qwen3.8 Max](https://meirinho.com.br/en/ia/modelos/qwen3-8-max/) | 65.6% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.8 Max |
| [Claude Sonnet 5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5/) | 65.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Sonnet 5 |
| [Qwen3.7 Max](https://meirinho.com.br/en/ia/modelos/qwen3-7-max/) | 64.6% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.7 Max |
| [Gemini 3.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash/) | 62.8% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3.5 Flash |
| [Gemini 3.1 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-1-pro/) | 59.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3.1 Pro |
| [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) | 59.2% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GLM-5.2 |
| [Gemini 3.6 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-6-flash/) | 58.9% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3.6 Flash |
| [DeepSeek V4 Flash 0731](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash-0731/) | 57.5% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | DeepSeek V4 Flash 0731 |
| [Kimi K2.6](https://meirinho.com.br/en/ia/modelos/kimi-k2-6/) | 57.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Kimi K2.6 |
| [Grok 4.5](https://meirinho.com.br/en/ia/modelos/grok-4-5/) | 57.2% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Grok 4.5 |
| [GLM-5.3-Flash](https://meirinho.com.br/en/ia/modelos/glm-5-3-flash/) | 55.8% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GLM-5.3-Flash |
| [GPT-5 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-pro/) | 55.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5 Pro |
| [GPT-5](https://meirinho.com.br/en/ia/modelos/gpt-5/) | 55.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5 |
| [Kimi K2.7 Code](https://meirinho.com.br/en/ia/modelos/kimi-k2-7-code/) | 54% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Kimi K2.7 Code |
| [Grok 4.7](https://meirinho.com.br/en/ia/modelos/grok-4-7/) | 53% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Grok 4.7 |
| [GPT-5.4 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-4-mini/) | 51.2% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.4 Mini |
| [Gemini 3 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-flash/) | 51.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3 Flash |
| [GPT-5 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-mini/) | 46.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5 Mini |
| [DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/deepseek-v4-pro/) | 45.3% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | DeepSeek V4 Pro |
| [GPT-5.4 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-4-nano/) | 44.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.4 Nano |
| [Grok 4.20](https://meirinho.com.br/en/ia/modelos/grok-4-20/) | 44.9% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Grok 4.20 |
| [Grok 4.3 Beta](https://meirinho.com.br/en/ia/modelos/grok-4-3-beta/) | 42.8% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Grok 4.3 Beta |
| [Qwen3.6 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-6-plus/) | 38.2% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.6 Plus |
| [GLM 5.1](https://meirinho.com.br/en/ia/modelos/glm-5-1/) | 36.8% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GLM 5.1 |
| [o4-mini](https://meirinho.com.br/en/ia/modelos/o4-mini/) | 36.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | o4-mini |
| [Qwen3.6 27B](https://meirinho.com.br/en/ia/modelos/qwen3-6-27b/) | 35.1% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.6 27B |
| [Claude Opus 4.5](https://meirinho.com.br/en/ia/modelos/claude-opus-4-5/) | 34.4% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Opus 4.5 |
| [Qwen3.7 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-7-plus/) | 34.4% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.7 Plus |
| [o3](https://meirinho.com.br/en/ia/modelos/o3/) | 33.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | o3 |
| [Qwen3.5 397B A17B](https://meirinho.com.br/en/ia/modelos/qwen3-5-397b-a17b/) | 31.2% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.5 397B A17B |
| [Gemini 3.1 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-1-flash-lite/) | 27.7% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3.1 Flash Lite |
| [GPT-5.5 Instant](https://meirinho.com.br/en/ia/modelos/gpt-5-5-instant/) | 26.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5.5 Instant |
| [Gemini 3.5 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash-lite/) | 26% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 3.5 Flash Lite |
| [Gemini 2.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-2-5-pro/) | 24.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Gemini 2.5 Pro |
| [Claude Sonnet 4.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-5/) | 23.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Sonnet 4.5 |
| [Qwen3.6 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-6-flash/) | 22.5% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.6 Flash |
| [Qwen3.6 35B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-6-35b-a3b/) | 20.4% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.6 35B A3B |
| [GPT-5 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-nano/) | 20% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-5 Nano |
| [Qwen3.7 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-7-flash/) | 19.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.7 Flash |
| [Qwen3-Max](https://meirinho.com.br/en/ia/modelos/qwen3-max/) | 18.9% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3-Max |
| [o3-mini](https://meirinho.com.br/en/ia/modelos/o3-mini/) | 18.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | o3-mini |
| [Qwen3.5-Flash](https://meirinho.com.br/en/ia/modelos/qwen3-5-flash-02-23/) | 18.2% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Qwen3.5-Flash |
| [o1](https://meirinho.com.br/en/ia/modelos/o1/) | 14.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | o1 |
| [Claude Opus 4.1](https://meirinho.com.br/en/ia/modelos/claude-opus-4-1/) | 12.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | Claude Opus 4.1 |
| [GPT-4.1 Mini](https://meirinho.com.br/en/ia/modelos/gpt-4-1-mini/) | 6.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-4.1 Mini |
| [GPT-4.1](https://meirinho.com.br/en/ia/modelos/gpt-4-1/) | 6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-4.1 |
| [GPT-4o-mini](https://meirinho.com.br/en/ia/modelos/gpt-4o-mini/) | 0.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-4o-mini |
| [GPT-4 Turbo (Apr 2024)](https://meirinho.com.br/en/ia/modelos/gpt-4-turbo-apr-2024/) | 0.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/frontiermath-tiers-1-3-v2) | GPT-4 Turbo (Apr 2024) |

## [SimpleQA Verified](https://meirinho.com.br/en/ia/modelos/benchmark/simpleqa-verified/)

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/texto/)

| Model | % | Lab | Weights | Source | Compare models |
| --- | --- | --- | --- | --- | --- |
| [Claude Fable 5.2](https://meirinho.com.br/en/ia/modelos/claude-fable-5-2/) | 94% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Fable 5.2 |
| [GPT-6.2 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-2-astra/) | 78.5% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-6.2 Astra |
| [Gemini 4 Carbon](https://meirinho.com.br/en/ia/modelos/gemini-4-carbon/) | 77.7% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 4 Carbon |
| [Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/gemini-4-argon/) | 76% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 4 Argon |
| [GPT-6 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-astra/) | 75.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-6 Astra |
| [GPT-6.1 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-1-sol/) | 73.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-6.1 Sol |
| [Gemini 3.1 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-1-pro/) | 73.5% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 3.1 Pro |
| [Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/claude-opus-5-5/) | 72.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Opus 5.5 |
| [Claude Fable 5.1](https://meirinho.com.br/en/ia/modelos/claude-fable-5-1/) | 70.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Fable 5.1 |
| [Claude Fable 5](https://meirinho.com.br/en/ia/modelos/claude-fable-5/) | 70.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Fable 5 |
| [Gemini 3.8 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-8-flash/) | 69.7% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 3.8 Flash |
| [GPT-5.6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol/) | 69.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.6 Sol |
| [Gemini 3.7 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-7-flash/) | 69.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 3.7 Flash |
| [Gemini 3 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-flash/) | 66.8% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 3 Flash |
| [Gemini 3.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash/) | 66.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 3.5 Flash |
| [Gemini 3.6 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-6-flash/) | 66.2% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Gemini 3.6 Flash |
| [GPT-5.5](https://meirinho.com.br/en/ia/modelos/gpt-5-5/) | 63% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.5 |
| [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) | 62.2% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Kimi K3.1 Pro |
| [GPT-6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-sol/) | 60.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-6 Sol |
| [Muse Spark 1.2](https://meirinho.com.br/en/ia/modelos/muse-spark-1-2/) | 60.3% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Muse Spark 1.2 |
| [Claude Opus 5](https://meirinho.com.br/en/ia/modelos/claude-opus-5/) | 59.9% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Opus 5 |
| [Muse Spark 1.1](https://meirinho.com.br/en/ia/modelos/muse-spark-1-1/) | 57.8% Epoch AI | Meta | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Muse Spark 1.1 |
| [GPT-6.1 Terra](https://meirinho.com.br/en/ia/modelos/gpt-6-1-terra/) | 57.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-6.1 Terra |
| [Muse Spark Open](https://meirinho.com.br/en/ia/modelos/muse-spark-open/) | 56.6% Epoch AI | Meta | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Muse Spark Open |
| [Grok 4.7](https://meirinho.com.br/en/ia/modelos/grok-4-7/) | 56% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Grok 4.7 |
| [Qwen3.7 Max](https://meirinho.com.br/en/ia/modelos/qwen3-7-max/) | 55.8% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3.7 Max |
| [Claude Opus 4.8](https://meirinho.com.br/en/ia/modelos/claude-opus-4-8/) | 53% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Opus 4.8 |
| [Qwen 3.6 Max (Preview)](https://meirinho.com.br/en/ia/modelos/qwen-3-6-max-preview/) | 52% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen 3.6 Max (Preview) |
| [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) | 51.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Opus 4.7 |
| [Kimi K3](https://meirinho.com.br/en/ia/modelos/kimi-k3/) | 50.6% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Kimi K3 |
| [GPT-5](https://meirinho.com.br/en/ia/modelos/gpt-5/) | 50.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5 |
| [o3](https://meirinho.com.br/en/ia/modelos/o3/) | 49.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | o3 |
| [Grok 4.6](https://meirinho.com.br/en/ia/modelos/grok-4-6/) | 49.3% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Grok 4.6 |
| [Qwen3-Max](https://meirinho.com.br/en/ia/modelos/qwen3-max/) | 48.7% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3-Max |
| [Grok 4.5](https://meirinho.com.br/en/ia/modelos/grok-4-5/) | 48.3% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Grok 4.5 |
| [GPT-5.1](https://meirinho.com.br/en/ia/modelos/gpt-5-1/) | 48% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.1 |
| [Qwen3.8 Max](https://meirinho.com.br/en/ia/modelos/qwen3-8-max/) | 47.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3.8 Max |
| [DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/deepseek-v4-pro/) | 47% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | DeepSeek V4 Pro |
| [Claude Opus 4.6](https://meirinho.com.br/en/ia/modelos/claude-opus-4-6/) | 47% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Opus 4.6 |
| [Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5-5/) | 46.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Sonnet 5.5 |
| [GPT-5.4 Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-4-pro/) | 46.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.4 Pro |
| [Claude Opus 4.5](https://meirinho.com.br/en/ia/modelos/claude-opus-4-5/) | 45.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Opus 4.5 |
| [GPT-5.4](https://meirinho.com.br/en/ia/modelos/gpt-5-4/) | 45.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.4 |
| [Doubao Seed 2.1 Pro](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-1-pro/) | 44.7% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Doubao Seed 2.1 Pro |
| [Qwen3.6 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-6-plus/) | 44.1% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3.6 Plus |
| [GPT-5.6 Terra](https://meirinho.com.br/en/ia/modelos/gpt-5-6-terra/) | 43.2% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.6 Terra |
| [DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/deepseek-v4-1-flash/) | 42.9% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | DeepSeek V4.1 Flash |
| [Doubao Seed Evolving (Nov 2026)](https://meirinho.com.br/en/ia/modelos/doubao-seed-evolving-nov-2026/) | 41.9% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Doubao Seed Evolving (Nov 2026) |
| [GPT-6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-6-luna/) | 41.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-6 Luna |
| [o1](https://meirinho.com.br/en/ia/modelos/o1/) | 41.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | o1 |
| [GLM-5.3](https://meirinho.com.br/en/ia/modelos/glm-5-3/) | 41% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GLM-5.3 |
| [GPT-5.6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna/) | 41% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.6 Luna |
| [Qwen3-235B-A22B](https://meirinho.com.br/en/ia/modelos/qwen3-235b-a22b/) | 40.4% Epoch AI | Alibaba | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3-235B-A22B |
| [Doubao-Seed-2.0 Pro](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-0-pro/) | 39.4% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Doubao-Seed-2.0 Pro |
| [GPT-5.2](https://meirinho.com.br/en/ia/modelos/gpt-5-2/) | 37.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.2 |
| [Kimi K2.7 Code](https://meirinho.com.br/en/ia/modelos/kimi-k2-7-code/) | 36.5% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Kimi K2.7 Code |
| [Claude Sonnet 4.6](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-6/) | 35.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Sonnet 4.6 |
| [Kimi K2.6](https://meirinho.com.br/en/ia/modelos/kimi-k2-6/) | 34.9% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Kimi K2.6 |
| [Kimi K2.5](https://meirinho.com.br/en/ia/modelos/kimi-k2-5/) | 34.3% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Kimi K2.5 |
| [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) | 34.2% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GLM-5.2 |
| [GLM 5.1](https://meirinho.com.br/en/ia/modelos/glm-5-1/) | 34% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GLM 5.1 |
| [Claude Sonnet 5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5/) | 33.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Sonnet 5 |
| [DeepSeek V4 Flash 0731](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash-0731/) | 33.6% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | DeepSeek V4 Flash 0731 |
| [Grok 4.3 Beta](https://meirinho.com.br/en/ia/modelos/grok-4-3-beta/) | 33.2% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Grok 4.3 Beta |
| [GLM 4.7](https://meirinho.com.br/en/ia/modelos/glm-4-7/) | 32.2% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GLM 4.7 |
| [Doubao Seed 2.0 Lite (Apr 2026)](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-0-lite-apr-2026/) | 31.5% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Doubao Seed 2.0 Lite (Apr 2026) |
| [GPT-4.1](https://meirinho.com.br/en/ia/modelos/gpt-4-1/) | 31.1% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-4.1 |
| [Claude Sonnet 4.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-5/) | 30.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Sonnet 4.5 |
| [Grok 4.20](https://meirinho.com.br/en/ia/modelos/grok-4-20/) | 30.2% Epoch AI | xAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Grok 4.20 |
| [Doubao Seed 2.1 Turbo](https://meirinho.com.br/en/ia/modelos/doubao-seed-2-1-turbo/) | 29.7% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Doubao Seed 2.1 Turbo |
| [GPT-5.4 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-4-mini/) | 29.4% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.4 Mini |
| [Qwen3.5 Plus 2026-02-15](https://meirinho.com.br/en/ia/modelos/qwen3-5-plus-02-15/) | 25.4% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3.5 Plus 2026-02-15 |
| [Claude Haiku 5.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-5-5/) | 23.8% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Haiku 5.5 |
| [Mistral Large 4](https://meirinho.com.br/en/ia/modelos/mistral-large-4/) | 22.4% Epoch AI | Mistral AI | open | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Mistral Large 4 |
| [GPT-5 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-mini/) | 21.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5 Mini |
| [Qwen3.5-Flash](https://meirinho.com.br/en/ia/modelos/qwen3-5-flash-02-23/) | 20.3% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3.5-Flash |
| [o4-mini](https://meirinho.com.br/en/ia/modelos/o4-mini/) | 19.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | o4-mini |
| [Doubao-Seed-1.6](https://meirinho.com.br/en/ia/modelos/doubao-seed-1-6/) | 17.4% Epoch AI | ByteDance | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Doubao-Seed-1.6 |
| [Qwen3.6 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-6-flash/) | 15.9% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Qwen3.6 Flash |
| [o3-mini](https://meirinho.com.br/en/ia/modelos/o3-mini/) | 15.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | o3-mini |
| [Claude Haiku 4.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-4-5/) | 13.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude Haiku 4.5 |
| [GPT-4.1 Mini](https://meirinho.com.br/en/ia/modelos/gpt-4-1-mini/) | 12.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-4.1 Mini |
| [Claude 3 Opus](https://meirinho.com.br/en/ia/modelos/claude-3-opus/) | 12.6% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | Claude 3 Opus |
| [GPT-5 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-nano/) | 11.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5 Nano |
| [GPT-5.4 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-4-nano/) | 11.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-5.4 Nano |
| [GPT-4o-mini](https://meirinho.com.br/en/ia/modelos/gpt-4o-mini/) | 8.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-4o-mini |
| [GPT-4.1 Nano](https://meirinho.com.br/en/ia/modelos/gpt-4-1-nano/) | 6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/simpleqa-verified) | GPT-4.1 Nano |

## [SWE-bench Verified](https://meirinho.com.br/en/ia/modelos/benchmark/swe-bench-verified/)

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/codigo/)

| Model | % | Lab | Weights | Source | Compare models |
| --- | --- | --- | --- | --- | --- |
| [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) | 83.5% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude Opus 4.7 |
| [GPT-5.5](https://meirinho.com.br/en/ia/modelos/gpt-5-5/) | 80.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-5.5 |
| [Gemini 3.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash/) | 79.3% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Gemini 3.5 Flash |
| [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) | 78.7% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GLM-5.2 |
| [Claude Opus 4.6](https://meirinho.com.br/en/ia/modelos/claude-opus-4-6/) | 78.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude Opus 4.6 |
| [DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/deepseek-v4-pro/) | 77.6% Epoch AI | DeepSeek | open | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | DeepSeek V4 Pro |
| [Qwen3.7 Max](https://meirinho.com.br/en/ia/modelos/qwen3-7-max/) | 77.3% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Qwen3.7 Max |
| [GPT-5.4](https://meirinho.com.br/en/ia/modelos/gpt-5-4/) | 76.9% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-5.4 |
| [Kimi K2.6](https://meirinho.com.br/en/ia/modelos/kimi-k2-6/) | 76.7% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Kimi K2.6 |
| [Claude Opus 4.5](https://meirinho.com.br/en/ia/modelos/claude-opus-4-5/) | 76.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude Opus 4.5 |
| [Qwen 3.6 Max (Preview)](https://meirinho.com.br/en/ia/modelos/qwen-3-6-max-preview/) | 76.7% Epoch AI | Alibaba | undisclosed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Qwen 3.6 Max (Preview) |
| [Gemini 3.1 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-1-pro/) | 75.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Gemini 3.1 Pro |
| [Gemini 3 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-flash/) | 75.4% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Gemini 3 Flash |
| [Claude Sonnet 4.6](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-6/) | 75.2% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude Sonnet 4.6 |
| [GPT-5.3-Codex](https://meirinho.com.br/en/ia/modelos/gpt-5-3-codex/) | 74.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-5.3-Codex |
| [GLM 5.1](https://meirinho.com.br/en/ia/modelos/glm-5-1/) | 74.2% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GLM 5.1 |
| [GPT-5.2](https://meirinho.com.br/en/ia/modelos/gpt-5-2/) | 73.8% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-5.2 |
| [Kimi K2.5](https://meirinho.com.br/en/ia/modelos/kimi-k2-5/) | 73.8% Epoch AI | Moonshot AI | open | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Kimi K2.5 |
| [GPT-5](https://meirinho.com.br/en/ia/modelos/gpt-5/) | 73.6% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-5 |
| [Claude Opus 4.1](https://meirinho.com.br/en/ia/modelos/claude-opus-4-1/) | 73.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude Opus 4.1 |
| [Gemini 3 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-pro/) | 72.9% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Gemini 3 Pro |
| [GLM 5](https://meirinho.com.br/en/ia/modelos/glm-5/) | 72.1% Epoch AI | Zhipu AI (Z.ai) | open | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GLM 5 |
| [Claude Sonnet 4.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-5/) | 71.3% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude Sonnet 4.5 |
| [Claude Opus 4](https://meirinho.com.br/en/ia/modelos/claude-opus-4/) | 70.7% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude Opus 4 |
| [GPT-5.1](https://meirinho.com.br/en/ia/modelos/gpt-5-1/) | 68% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-5.1 |
| [GPT-5 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-mini/) | 64.7% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-5 Mini |
| [o3](https://meirinho.com.br/en/ia/modelos/o3/) | 62.3% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | o3 |
| [Claude 3.7 Sonnet](https://meirinho.com.br/en/ia/modelos/claude-3-7-sonnet/) | 61% Epoch AI | Anthropic | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Claude 3.7 Sonnet |
| [Qwen3.6 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-6-plus/) | 57.9% Epoch AI | Alibaba | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Qwen3.6 Plus |
| [Gemini 2.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-2-5-pro/) | 57.6% Epoch AI | Google | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | Gemini 2.5 Pro |
| [GPT-4.1](https://meirinho.com.br/en/ia/modelos/gpt-4-1/) | 48.5% Epoch AI | OpenAI | closed | [Epoch AI](https://epoch.ai/benchmarks/swe-bench-verified) | GPT-4.1 |

## [GPQA Diamond (OpenRouter)](https://meirinho.com.br/en/ia/modelos/benchmark/gpqa-diamond/)

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/texto/)

| Model | % | Lab | Weights | Source | Compare models |
| --- | --- | --- | --- | --- | --- |
| [Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/gemini-4-argon/) | 96.5% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 4 Argon |
| [Gemini 3.8 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-8-flash/) | 95.6% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3.8 Flash |
| [GPT-6 Astra Pro](https://meirinho.com.br/en/ia/modelos/gpt-6-astra-pro/) | 95.5% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-6 Astra Pro |
| [Gemini 3.1 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-1-pro/) | 95.2% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3.1 Pro |
| [GPT-6.1 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-1-sol/) | 94.8% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-6.1 Sol |
| [Ling 3.1 Flash](https://meirinho.com.br/en/ia/modelos/ling-3-1-flash/) | 94.8% OpenRouter Benchmarks | Ant Group | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Ling 3.1 Flash |
| [GPT-6 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-astra/) | 94.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-6 Astra |
| [Gemini 3.7 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-7-flash/) | 94.3% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3.7 Flash |
| [GPT-5.6 Sol Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol-pro/) | 94.1% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.6 Sol Pro |
| [GPT-5.5](https://meirinho.com.br/en/ia/modelos/gpt-5-5/) | 93.9% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.5 |
| [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) | 93.9% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Kimi K3.1 Pro |
| [Kimi K3](https://meirinho.com.br/en/ia/modelos/kimi-k3/) | 93.5% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Kimi K3 |
| [Gemini 3.6 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-6-flash/) | 93.4% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3.6 Flash |
| [Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5-5/) | 93% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Sonnet 5.5 |
| [DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/deepseek-v4-1-flash/) | 92.8% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V4.1 Flash |
| [Grok 4.6](https://meirinho.com.br/en/ia/modelos/grok-4-6/) | 92.8% OpenRouter Benchmarks | xAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Grok 4.6 |
| [GPT-5.6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol/) | 92.8% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.6 Sol |
| [Gemini 3.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash/) | 92.6% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3.5 Flash |
| [GLM-5.3-Flash](https://meirinho.com.br/en/ia/modelos/glm-5-3-flash/) | 92.5% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM-5.3-Flash |
| [GPT-6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-sol/) | 92.5% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-6 Sol |
| [Nemotron 3.5 Super VL](https://meirinho.com.br/en/ia/modelos/nemotron-3-5-super-vl/) | 92.2% OpenRouter Benchmarks | NVIDIA | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Nemotron 3.5 Super VL |
| [DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/deepseek-v4-pro/) | 91.9% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V4 Pro |
| [Grok 4.7](https://meirinho.com.br/en/ia/modelos/grok-4-7/) | 91.9% OpenRouter Benchmarks | xAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Grok 4.7 |
| [GPT-6.1 Terra](https://meirinho.com.br/en/ia/modelos/gpt-6-1-terra/) | 91.9% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-6.1 Terra |
| [Qwen3.8 2.4T A95B](https://meirinho.com.br/en/ia/modelos/qwen3-8-2-4t-a95b/) | 91.8% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.8 2.4T A95B |
| [MiniMax M3](https://meirinho.com.br/en/ia/modelos/minimax-m3/) | 91.3% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiniMax M3 |
| [Muse Spark Open](https://meirinho.com.br/en/ia/modelos/muse-spark-open/) | 91% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Muse Spark Open |
| [GPT-5.6 Luna Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna-pro/) | 91% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.6 Luna Pro |
| [DeepSeek V4 Flash Vision Exp](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash-vision-exp/) | 90.9% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V4 Flash Vision Exp |
| [Kimi K2.7 Code](https://meirinho.com.br/en/ia/modelos/kimi-k2-7-code/) | 90.9% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Kimi K2.7 Code |
| [Qwen3.7 Max](https://meirinho.com.br/en/ia/modelos/qwen3-7-max/) | 90.9% OpenRouter Benchmarks | Alibaba | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.7 Max |
| [Hy4](https://meirinho.com.br/en/ia/modelos/hy4/) | 90.7% OpenRouter Benchmarks | Tencent | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Hy4 |
| [MiMo V2.6 Pro](https://meirinho.com.br/en/ia/modelos/mimo-v2-6-pro/) | 90.6% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiMo V2.6 Pro |
| [Qwen3.8 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-8-flash/) | 90.5% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.8 Flash |
| [Gemini 3 Flash Preview](https://meirinho.com.br/en/ia/modelos/gemini-3-flash-preview/) | 90.4% OpenRouter Benchmarks | Google | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3 Flash Preview |
| [GPT-5.4](https://meirinho.com.br/en/ia/modelos/gpt-5-4/) | 90.3% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.4 |
| [Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/claude-opus-5-5/) | 90.2% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Opus 5.5 |
| [Kimi K2.5](https://meirinho.com.br/en/ia/modelos/kimi-k2-5/) | 90.2% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Kimi K2.5 |
| [Claude Opus 4.8](https://meirinho.com.br/en/ia/modelos/claude-opus-4-8/) | 90.1% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Opus 4.8 |
| [DeepSeek V4 Flash 0423](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash/) | 90.1% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V4 Flash 0423 |
| [Claude Fable 5.1](https://meirinho.com.br/en/ia/modelos/claude-fable-5-1/) | 89.9% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Fable 5.1 |
| [Claude Opus 5](https://meirinho.com.br/en/ia/modelos/claude-opus-5/) | 89.9% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Opus 5 |
| [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) | 89.8% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Opus 4.7 |
| [Kimi K2.6](https://meirinho.com.br/en/ia/modelos/kimi-k2-6/) | 89.7% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Kimi K2.6 |
| [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) | 89.7% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM-5.2 |
| [MiniMax M3.1 Flash](https://meirinho.com.br/en/ia/modelos/minimax-m3-1-flash/) | 89.3% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiniMax M3.1 Flash |
| [GPT-5.6 Terra](https://meirinho.com.br/en/ia/modelos/gpt-5-6-terra/) | 89.3% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.6 Terra |
| [GPT-6 Luna Pro](https://meirinho.com.br/en/ia/modelos/gpt-6-luna-pro/) | 89% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-6 Luna Pro |
| [Nova Micro 1.0](https://meirinho.com.br/en/ia/modelos/nova-micro-v1/) | 88.9% OpenRouter Benchmarks | Amazon | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Nova Micro 1.0 |
| [GPT-5.6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna/) | 88.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.6 Luna |
| [GPT-6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-6-luna/) | 88.2% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-6 Luna |
| [Qwen3.5 397B A17B](https://meirinho.com.br/en/ia/modelos/qwen3-5-397b-a17b/) | 88.1% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.5 397B A17B |
| [GLM-5.3](https://meirinho.com.br/en/ia/modelos/glm-5-3/) | 87.9% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM-5.3 |
| [Qwen3.7 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-7-plus/) | 87.9% OpenRouter Benchmarks | Alibaba | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.7 Plus |
| [Claude Fable 5](https://meirinho.com.br/en/ia/modelos/claude-fable-5/) | 87.4% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Fable 5 |
| [Qwen3.8 27B](https://meirinho.com.br/en/ia/modelos/qwen3-8-27b/) | 87.2% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.8 27B |
| [Claude Haiku 5.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-5-5/) | 87.1% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Haiku 5.5 |
| [GPT-5.1](https://meirinho.com.br/en/ia/modelos/gpt-5-1/) | 87% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.1 |
| [Claude Sonnet 5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5/) | 86.9% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Sonnet 5 |
| [MiMo-V2.5-Pro](https://meirinho.com.br/en/ia/modelos/mimo-v2-5-pro/) | 86.5% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiMo-V2.5-Pro |
| [Claude Opus 4.5](https://meirinho.com.br/en/ia/modelos/claude-opus-4-5/) | 86.4% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Opus 4.5 |
| [Claude Opus 4.6](https://meirinho.com.br/en/ia/modelos/claude-opus-4-6/) | 86.4% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Opus 4.6 |
| [MiniMax M2.7](https://meirinho.com.br/en/ia/modelos/minimax-m2-7/) | 86.1% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiniMax M2.7 |
| [Gemma 4 31B](https://meirinho.com.br/en/ia/modelos/gemma-4-31b/) | 85.9% OpenRouter Benchmarks | Google | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemma 4 31B |
| [GPT-5](https://meirinho.com.br/en/ia/modelos/gpt-5/) | 85.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5 |
| [MiMo-V2.6-Flash](https://meirinho.com.br/en/ia/modelos/mimo-v2-6-flash/) | 85.2% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiMo-V2.6-Flash |
| [Qwen3.6 27B](https://meirinho.com.br/en/ia/modelos/qwen3-6-27b/) | 85.1% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.6 27B |
| [Qwen3.6 35B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-6-35b-a3b/) | 85.1% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.6 35B A3B |
| [MiniMax M2.5](https://meirinho.com.br/en/ia/modelos/minimax-m2-5/) | 85% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiniMax M2.5 |
| [MiMo-V2.5](https://meirinho.com.br/en/ia/modelos/mimo-v2-5/) | 84.9% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | MiMo-V2.5 |
| [Ling 3.0 Flash VL](https://meirinho.com.br/en/ia/modelos/ling-3-0-flash-vl/) | 84.8% OpenRouter Benchmarks | Ant Group | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Ling 3.0 Flash VL |
| [Qwen3.5-122B-A10B](https://meirinho.com.br/en/ia/modelos/qwen3-5-122b-a10b/) | 84.8% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.5-122B-A10B |
| [Gemini 3.5 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash-lite/) | 84.4% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3.5 Flash Lite |
| [GLM 4.6](https://meirinho.com.br/en/ia/modelos/glm-4-6/) | 84.3% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM 4.6 |
| [GPT-5.4 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-4-mini/) | 84.2% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.4 Mini |
| [GLM 5.1](https://meirinho.com.br/en/ia/modelos/glm-5-1/) | 84.1% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM 5.1 |
| [GLM 4.7](https://meirinho.com.br/en/ia/modelos/glm-4-7/) | 83.8% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM 4.7 |
| [DeepSeek V3.2](https://meirinho.com.br/en/ia/modelos/deepseek-v3-2/) | 83% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V3.2 |
| [Qwen3.5-35B-A3B](https://meirinho.com.br/en/ia/modelos/qwen3-5-35b-a3b/) | 82.9% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.5-35B-A3B |
| [Gemini 2.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-2-5-pro/) | 82.6% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 2.5 Pro |
| [DeepSeek V3.1 Terminus](https://meirinho.com.br/en/ia/modelos/deepseek-v3-1-terminus/) | 82.4% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V3.1 Terminus |
| [Gemini 3.1 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-1-flash-lite/) | 82.1% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 3.1 Flash Lite |
| [Muse Glimmer 30B](https://meirinho.com.br/en/ia/modelos/muse-glimmer-30b/) | 82.1% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Muse Glimmer 30B |
| [Claude Sonnet 4.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-5/) | 81.7% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Sonnet 4.5 |
| [DeepSeek V3.1](https://meirinho.com.br/en/ia/modelos/deepseek-chat-v3-1/) | 81.7% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V3.1 |
| [GPT-5 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-mini/) | 81% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5 Mini |
| [Qwen3.5-9B](https://meirinho.com.br/en/ia/modelos/qwen3-5-9b/) | 81% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3.5-9B |
| [GLM 5](https://meirinho.com.br/en/ia/modelos/glm-5/) | 81% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM 5 |
| [DeepSeek R1](https://meirinho.com.br/en/ia/modelos/deepseek-r1/) | 79.7% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek R1 |
| [Qwen3 235B A22B Thinking 2507](https://meirinho.com.br/en/ia/modelos/qwen3-235b-a22b-thinking-2507/) | 79.1% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 235B A22B Thinking 2507 |
| [Gemma 4 26B A4B](https://meirinho.com.br/en/ia/modelos/gemma-4-26b-a4b-it/) | 78.8% OpenRouter Benchmarks | Google | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemma 4 26B A4B |
| [GPT-5.4 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-4-nano/) | 77.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5.4 Nano |
| [gpt-oss-120b](https://meirinho.com.br/en/ia/modelos/gpt-oss-120b/) | 77.4% OpenRouter Benchmarks | OpenAI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | gpt-oss-120b |
| [Qwen3-235B-A22B](https://meirinho.com.br/en/ia/modelos/qwen3-235b-a22b/) | 77.4% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3-235B-A22B |
| [Gemini 2.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-2-5-flash/) | 76.7% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 2.5 Flash |
| [Ling 3.0 Flash](https://meirinho.com.br/en/ia/modelos/ling-3-0-flash/) | 76.5% OpenRouter Benchmarks | Ant Group | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Ling 3.0 Flash |
| [Qwen3 Coder Next](https://meirinho.com.br/en/ia/modelos/qwen3-coder-next/) | 76.3% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 Coder Next |
| [Mistral Small 4](https://meirinho.com.br/en/ia/modelos/mistral-small-2603/) | 75.8% OpenRouter Benchmarks | Mistral AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Mistral Small 4 |
| [Nemotron 3.5 Lightning](https://meirinho.com.br/en/ia/modelos/nemotron-3-5-lightning/) | 74.8% OpenRouter Benchmarks | NVIDIA | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Nemotron 3.5 Lightning |
| [Claude Haiku 4.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-4-5/) | 73% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Haiku 4.5 |
| [Qwen3 Next 80B A3B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-next-80b-a3b-instruct/) | 71.8% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 Next 80B A3B Instruct |
| [Qwen3 VL 235B A22B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-vl-235b-a22b-instruct/) | 71.1% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 VL 235B A22B Instruct |
| [Claude Sonnet 4.6](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-6/) | 70.6% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Claude Sonnet 4.6 |
| [GPT-5 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-nano/) | 70.5% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-5 Nano |
| [Nemotron 3 Nano 30B A3B](https://meirinho.com.br/en/ia/modelos/nemotron-3-nano-30b-a3b/) | 68.8% OpenRouter Benchmarks | NVIDIA | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Nemotron 3 Nano 30B A3B |
| [gpt-oss-20b](https://meirinho.com.br/en/ia/modelos/gpt-oss-20b/) | 68.7% OpenRouter Benchmarks | OpenAI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | gpt-oss-20b |
| [Qwen3 VL 30B A3B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-vl-30b-a3b-instruct/) | 66.8% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 VL 30B A3B Instruct |
| [GLM 4.6V](https://meirinho.com.br/en/ia/modelos/glm-4-6v/) | 66.8% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM 4.6V |
| [DeepSeek V3](https://meirinho.com.br/en/ia/modelos/deepseek-v3/) | 66.7% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | DeepSeek V3 |
| [Llama 4 Maverick](https://meirinho.com.br/en/ia/modelos/llama-4-maverick/) | 66.7% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Llama 4 Maverick |
| [GPT-4.1](https://meirinho.com.br/en/ia/modelos/gpt-4-1/) | 66% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-4.1 |
| [Granite 4.2 8B](https://meirinho.com.br/en/ia/modelos/granite-4-2-8b/) | 65.7% OpenRouter Benchmarks | IBM | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Granite 4.2 8B |
| [GPT-4.1 Mini](https://meirinho.com.br/en/ia/modelos/gpt-4-1-mini/) | 65.2% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-4.1 Mini |
| [Qwen3 30B A3B Instruct 2507](https://meirinho.com.br/en/ia/modelos/qwen3-30b-a3b-instruct-2507/) | 64.8% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 30B A3B Instruct 2507 |
| [Qwen3 32B](https://meirinho.com.br/en/ia/modelos/qwen3-32b/) | 63.3% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 32B |
| [Qwen3-Coder](https://meirinho.com.br/en/ia/modelos/qwen3-coder/) | 61.2% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3-Coder |
| [Qwen3 14B](https://meirinho.com.br/en/ia/modelos/qwen3-14b/) | 61.2% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 14B |
| [Qwen3 30B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-30b-a3b/) | 60.9% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 30B A3B |
| [Gemini 2.5 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-2-5-flash-lite/) | 59.6% OpenRouter Benchmarks | Google | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemini 2.5 Flash Lite |
| [GLM 4.7 Flash](https://meirinho.com.br/en/ia/modelos/glm-4-7-flash/) | 58% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GLM 4.7 Flash |
| [Qwen3 VL 8B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-vl-8b-instruct/) | 53.9% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 VL 8B Instruct |
| [Qwen3 Coder 30B A3B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-coder-30b-a3b-instruct/) | 53.4% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen3 Coder 30B A3B Instruct |
| [GPT-4o](https://meirinho.com.br/en/ia/modelos/gpt-4o/) | 52.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-4o |
| [GPT-4.1 Nano](https://meirinho.com.br/en/ia/modelos/gpt-4-1-nano/) | 50.5% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-4.1 Nano |
| [Llama 3.3 70B Instruct](https://meirinho.com.br/en/ia/modelos/llama-3-3-70b-instruct/) | 49.8% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Llama 3.3 70B Instruct |
| [Qwen2.5 72B Instruct](https://meirinho.com.br/en/ia/modelos/qwen-2-5-72b-instruct/) | 47.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen2.5 72B Instruct |
| [Qwen2.5 VL 72B Instruct](https://meirinho.com.br/en/ia/modelos/qwen2-5-vl-72b-instruct/) | 45.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Qwen2.5 VL 72B Instruct |
| [GPT-4o-mini](https://meirinho.com.br/en/ia/modelos/gpt-4o-mini/) | 43.6% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | GPT-4o-mini |
| [Gemma 3 12B](https://meirinho.com.br/en/ia/modelos/gemma-3-12b-it/) | 40.6% OpenRouter Benchmarks | Google | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemma 3 12B |
| [Hy-MT2-30B-A3B](https://meirinho.com.br/en/ia/modelos/hy-mt2-30b-a3b/) | 38.9% OpenRouter Benchmarks | Tencent | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Hy-MT2-30B-A3B |
| [Llama 3.1 8B Instruct](https://meirinho.com.br/en/ia/modelos/llama-3-1-8b-instruct/) | 29% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Llama 3.1 8B Instruct |
| [Llama 3.2 3B Instruct](https://meirinho.com.br/en/ia/modelos/llama-3-2-3b-instruct/) | 28.6% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Llama 3.2 3B Instruct |
| [Gemma 3 4B](https://meirinho.com.br/en/ia/modelos/gemma-3-4b-it/) | 27.1% OpenRouter Benchmarks | Google | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Gemma 3 4B |
| [Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)](https://meirinho.com.br/en/ia/modelos/gemini-3-1-flash-lite-image/) | 11.6% OpenRouter Benchmarks | Google | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond) | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) |

## [τ-bench Verified (airline)](https://meirinho.com.br/en/ia/modelos/benchmark/tau-bench-airline/)

[How to read each metric](https://meirinho.com.br/en/ia/como-ler-benchmarks/agentes/)

| Model | % | Lab | Weights | Source | Compare models |
| --- | --- | --- | --- | --- | --- |
| [Claude Fable 5](https://meirinho.com.br/en/ia/modelos/claude-fable-5/) | 82% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Fable 5 |
| [DeepSeek V4 Flash 0423](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash/) | 81.7% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V4 Flash 0423 |
| [Gemini 3.7 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-7-flash/) | 80.7% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3.7 Flash |
| [Qwen3.5 397B A17B](https://meirinho.com.br/en/ia/modelos/qwen3-5-397b-a17b/) | 80.2% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.5 397B A17B |
| [GPT-5.1](https://meirinho.com.br/en/ia/modelos/gpt-5-1/) | 80% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.1 |
| [Ling 3.1 Flash](https://meirinho.com.br/en/ia/modelos/ling-3-1-flash/) | 79.3% OpenRouter Benchmarks | Ant Group | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Ling 3.1 Flash |
| [Claude Fable 5.1](https://meirinho.com.br/en/ia/modelos/claude-fable-5-1/) | 79.2% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Fable 5.1 |
| [Qwen3.6 27B](https://meirinho.com.br/en/ia/modelos/qwen3-6-27b/) | 79% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.6 27B |
| [MiMo V2.6 Pro](https://meirinho.com.br/en/ia/modelos/mimo-v2-6-pro/) | 78.9% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiMo V2.6 Pro |
| [Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5-5/) | 78.8% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Sonnet 5.5 |
| [GLM-5.3](https://meirinho.com.br/en/ia/modelos/glm-5-3/) | 78.8% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM-5.3 |
| [Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/gemini-4-argon/) | 78.7% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 4 Argon |
| [Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/kimi-k3-1-pro/) | 78.7% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Kimi K3.1 Pro |
| [Claude Opus 5](https://meirinho.com.br/en/ia/modelos/claude-opus-5/) | 78.7% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Opus 5 |
| [Qwen3.5-35B-A3B](https://meirinho.com.br/en/ia/modelos/qwen3-5-35b-a3b/) | 78.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.5-35B-A3B |
| [DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/deepseek-v4-pro/) | 78.6% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V4 Pro |
| [Gemini 3.5 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash-lite/) | 78.3% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3.5 Flash Lite |
| [MiMo-V2.6-Flash](https://meirinho.com.br/en/ia/modelos/mimo-v2-6-flash/) | 78.2% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiMo-V2.6-Flash |
| [Qwen3.8 27B](https://meirinho.com.br/en/ia/modelos/qwen3-8-27b/) | 78% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.8 27B |
| [DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/deepseek-v4-1-flash/) | 77.8% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V4.1 Flash |
| [Claude Opus 4.7](https://meirinho.com.br/en/ia/modelos/claude-opus-4-7/) | 77.8% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Opus 4.7 |
| [Qwen3.5-122B-A10B](https://meirinho.com.br/en/ia/modelos/qwen3-5-122b-a10b/) | 77.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.5-122B-A10B |
| [Hy4](https://meirinho.com.br/en/ia/modelos/hy4/) | 77.6% OpenRouter Benchmarks | Tencent | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Hy4 |
| [Claude Opus 4.8](https://meirinho.com.br/en/ia/modelos/claude-opus-4-8/) | 77.5% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Opus 4.8 |
| [Qwen3.8 2.4T A95B](https://meirinho.com.br/en/ia/modelos/qwen3-8-2-4t-a95b/) | 77.4% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.8 2.4T A95B |
| [Kimi K2.7 Code](https://meirinho.com.br/en/ia/modelos/kimi-k2-7-code/) | 77.3% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Kimi K2.7 Code |
| [Muse Glimmer 30B](https://meirinho.com.br/en/ia/modelos/muse-glimmer-30b/) | 77.2% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Muse Glimmer 30B |
| [Gemini 3.1 Pro](https://meirinho.com.br/en/ia/modelos/gemini-3-1-pro/) | 77% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3.1 Pro |
| [Gemma 4 31B](https://meirinho.com.br/en/ia/modelos/gemma-4-31b/) | 77% OpenRouter Benchmarks | Google | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemma 4 31B |
| [GLM 5](https://meirinho.com.br/en/ia/modelos/glm-5/) | 76.9% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM 5 |
| [Gemini 3.8 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-8-flash/) | 76.7% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3.8 Flash |
| [GLM-5.2](https://meirinho.com.br/en/ia/modelos/glm-5-2/) | 76.7% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM-5.2 |
| [Nova Micro 1.0](https://meirinho.com.br/en/ia/modelos/nova-micro-v1/) | 76.7% OpenRouter Benchmarks | Amazon | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Nova Micro 1.0 |
| [DeepSeek V4 Flash Vision Exp](https://meirinho.com.br/en/ia/modelos/deepseek-v4-flash-vision-exp/) | 76.7% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V4 Flash Vision Exp |
| [Claude Opus 4.5](https://meirinho.com.br/en/ia/modelos/claude-opus-4-5/) | 76.6% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Opus 4.5 |
| [Qwen3.8 Flash](https://meirinho.com.br/en/ia/modelos/qwen3-8-flash/) | 76.4% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.8 Flash |
| [GLM-5.3-Flash](https://meirinho.com.br/en/ia/modelos/glm-5-3-flash/) | 76.3% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM-5.3-Flash |
| [MiniMax M3](https://meirinho.com.br/en/ia/modelos/minimax-m3/) | 76.3% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiniMax M3 |
| [Gemini 3.6 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-6-flash/) | 76.3% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3.6 Flash |
| [Kimi K3](https://meirinho.com.br/en/ia/modelos/kimi-k3/) | 76.2% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Kimi K3 |
| [Qwen3.6 35B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-6-35b-a3b/) | 76.2% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.6 35B A3B |
| [GLM 4.7](https://meirinho.com.br/en/ia/modelos/glm-4-7/) | 76% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM 4.7 |
| [Claude Opus 4.6](https://meirinho.com.br/en/ia/modelos/claude-opus-4-6/) | 75.8% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Opus 4.6 |
| [GLM 5.1](https://meirinho.com.br/en/ia/modelos/glm-5-1/) | 75.8% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM 5.1 |
| [Muse Spark Open](https://meirinho.com.br/en/ia/modelos/muse-spark-open/) | 75.7% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Muse Spark Open |
| [Qwen3.5-9B](https://meirinho.com.br/en/ia/modelos/qwen3-5-9b/) | 75.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.5-9B |
| [Claude Sonnet 5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-5/) | 75.4% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Sonnet 5 |
| [Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/claude-opus-5-5/) | 75.3% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Opus 5.5 |
| [Grok 4.7](https://meirinho.com.br/en/ia/modelos/grok-4-7/) | 75.3% OpenRouter Benchmarks | xAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Grok 4.7 |
| [Grok 4.6](https://meirinho.com.br/en/ia/modelos/grok-4-6/) | 75.3% OpenRouter Benchmarks | xAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Grok 4.6 |
| [GPT-5.3-Codex](https://meirinho.com.br/en/ia/modelos/gpt-5-3-codex/) | 75.3% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.3-Codex |
| [GPT-5.4](https://meirinho.com.br/en/ia/modelos/gpt-5-4/) | 75% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.4 |
| [Claude Sonnet 4.5](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-5/) | 74.9% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Sonnet 4.5 |
| [Kimi K2.6](https://meirinho.com.br/en/ia/modelos/kimi-k2-6/) | 74.9% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Kimi K2.6 |
| [Nemotron 3.5 Super VL](https://meirinho.com.br/en/ia/modelos/nemotron-3-5-super-vl/) | 74.9% OpenRouter Benchmarks | NVIDIA | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Nemotron 3.5 Super VL |
| [Claude Sonnet 4.6](https://meirinho.com.br/en/ia/modelos/claude-sonnet-4-6/) | 74.8% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Sonnet 4.6 |
| [GPT-5.5](https://meirinho.com.br/en/ia/modelos/gpt-5-5/) | 74.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.5 |
| [DeepSeek V3.2](https://meirinho.com.br/en/ia/modelos/deepseek-v3-2/) | 74.7% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V3.2 |
| [MiniMax M3.1 Flash](https://meirinho.com.br/en/ia/modelos/minimax-m3-1-flash/) | 74.7% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiniMax M3.1 Flash |
| [MiMo-V2.5](https://meirinho.com.br/en/ia/modelos/mimo-v2-5/) | 74.7% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiMo-V2.5 |
| [GPT-5.6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol/) | 74.2% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.6 Sol |
| [MiMo-V2.5-Pro](https://meirinho.com.br/en/ia/modelos/mimo-v2-5-pro/) | 74.1% OpenRouter Benchmarks | Xiaomi | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiMo-V2.5-Pro |
| [Claude Haiku 5.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-5-5/) | 74% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Haiku 5.5 |
| [Ling 3.0 Flash](https://meirinho.com.br/en/ia/modelos/ling-3-0-flash/) | 74% OpenRouter Benchmarks | Ant Group | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Ling 3.0 Flash |
| [Kimi K2.5](https://meirinho.com.br/en/ia/modelos/kimi-k2-5/) | 73.8% OpenRouter Benchmarks | Moonshot AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Kimi K2.5 |
| [MiniMax M2.7](https://meirinho.com.br/en/ia/modelos/minimax-m2-7/) | 73.4% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiniMax M2.7 |
| [Gemini 3 Flash Preview](https://meirinho.com.br/en/ia/modelos/gemini-3-flash-preview/) | 73.3% OpenRouter Benchmarks | Google | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3 Flash Preview |
| [GLM 4.6](https://meirinho.com.br/en/ia/modelos/glm-4-6/) | 73.3% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM 4.6 |
| [DeepSeek V3.1](https://meirinho.com.br/en/ia/modelos/deepseek-chat-v3-1/) | 73.1% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V3.1 |
| [DeepSeek V3.1 Terminus](https://meirinho.com.br/en/ia/modelos/deepseek-v3-1-terminus/) | 73% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V3.1 Terminus |
| [Gemini 3.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-3-5-flash/) | 73% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3.5 Flash |
| [Gemini 3.1 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-3-1-flash-lite/) | 72.7% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 3.1 Flash Lite |
| [GPT-5.6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna/) | 72.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.6 Luna |
| [GPT-6 Astra Pro](https://meirinho.com.br/en/ia/modelos/gpt-6-astra-pro/) | 72.7% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-6 Astra Pro |
| [Qwen3.7 Plus](https://meirinho.com.br/en/ia/modelos/qwen3-7-plus/) | 72.7% OpenRouter Benchmarks | Alibaba | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.7 Plus |
| [GPT-5 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-mini/) | 72.4% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5 Mini |
| [GPT-5.6 Terra](https://meirinho.com.br/en/ia/modelos/gpt-5-6-terra/) | 72.3% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.6 Terra |
| [Ling 3.0 Flash VL](https://meirinho.com.br/en/ia/modelos/ling-3-0-flash-vl/) | 72% OpenRouter Benchmarks | Ant Group | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Ling 3.0 Flash VL |
| [Qwen3.7 Max](https://meirinho.com.br/en/ia/modelos/qwen3-7-max/) | 72% OpenRouter Benchmarks | Alibaba | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3.7 Max |
| [GPT-5.6 Sol Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-6-sol-pro/) | 71.7% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.6 Sol Pro |
| [Nemotron 3.5 Lightning](https://meirinho.com.br/en/ia/modelos/nemotron-3-5-lightning/) | 71.5% OpenRouter Benchmarks | NVIDIA | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Nemotron 3.5 Lightning |
| [GPT-6.1 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-1-sol/) | 70.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-6.1 Sol |
| [GPT-5](https://meirinho.com.br/en/ia/modelos/gpt-5/) | 70.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5 |
| [GPT-5.6 Luna Pro](https://meirinho.com.br/en/ia/modelos/gpt-5-6-luna-pro/) | 70.7% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.6 Luna Pro |
| [Gemma 4 26B A4B](https://meirinho.com.br/en/ia/modelos/gemma-4-26b-a4b-it/) | 70.6% OpenRouter Benchmarks | Google | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemma 4 26B A4B |
| [GPT-6 Sol](https://meirinho.com.br/en/ia/modelos/gpt-6-sol/) | 70.2% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-6 Sol |
| [GPT-6.1 Terra](https://meirinho.com.br/en/ia/modelos/gpt-6-1-terra/) | 70% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-6.1 Terra |
| [GPT-6 Astra](https://meirinho.com.br/en/ia/modelos/gpt-6-astra/) | 69.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-6 Astra |
| [GLM 4.7 Flash](https://meirinho.com.br/en/ia/modelos/glm-4-7-flash/) | 69.7% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM 4.7 Flash |
| [GPT-6 Luna Pro](https://meirinho.com.br/en/ia/modelos/gpt-6-luna-pro/) | 69.3% OpenRouter Benchmarks | OpenAI | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-6 Luna Pro |
| [MiniMax M2.5](https://meirinho.com.br/en/ia/modelos/minimax-m2-5/) | 69.2% OpenRouter Benchmarks | MiniMax | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | MiniMax M2.5 |
| [GPT-6 Luna](https://meirinho.com.br/en/ia/modelos/gpt-6-luna/) | 68.4% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-6 Luna |
| [gpt-oss-120b](https://meirinho.com.br/en/ia/modelos/gpt-oss-120b/) | 67.3% OpenRouter Benchmarks | OpenAI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | gpt-oss-120b |
| [GPT-5.4 Mini](https://meirinho.com.br/en/ia/modelos/gpt-5-4-mini/) | 66.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.4 Mini |
| [Claude Haiku 4.5](https://meirinho.com.br/en/ia/modelos/claude-haiku-4-5/) | 66.4% OpenRouter Benchmarks | Anthropic | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Claude Haiku 4.5 |
| [GPT-5.4 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-4-nano/) | 65.6% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5.4 Nano |
| [Granite 4.2 8B](https://meirinho.com.br/en/ia/modelos/granite-4-2-8b/) | 64% OpenRouter Benchmarks | IBM | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Granite 4.2 8B |
| [Qwen3 235B A22B Thinking 2507](https://meirinho.com.br/en/ia/modelos/qwen3-235b-a22b-thinking-2507/) | 61.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 235B A22B Thinking 2507 |
| [Qwen3 Coder Next](https://meirinho.com.br/en/ia/modelos/qwen3-coder-next/) | 60.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 Coder Next |
| [GLM 4.6V](https://meirinho.com.br/en/ia/modelos/glm-4-6v/) | 60.7% OpenRouter Benchmarks | Zhipu AI (Z.ai) | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GLM 4.6V |
| [Gemini 2.5 Pro](https://meirinho.com.br/en/ia/modelos/gemini-2-5-pro/) | 60% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 2.5 Pro |
| [Gemini 2.5 Flash](https://meirinho.com.br/en/ia/modelos/gemini-2-5-flash/) | 59.1% OpenRouter Benchmarks | Google | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 2.5 Flash |
| [DeepSeek R1](https://meirinho.com.br/en/ia/modelos/deepseek-r1/) | 58% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek R1 |
| [Qwen3 32B](https://meirinho.com.br/en/ia/modelos/qwen3-32b/) | 54.3% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 32B |
| [Gemini 2.5 Flash Lite](https://meirinho.com.br/en/ia/modelos/gemini-2-5-flash-lite/) | 53.7% OpenRouter Benchmarks | Google | undisclosed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemini 2.5 Flash Lite |
| [gpt-oss-20b](https://meirinho.com.br/en/ia/modelos/gpt-oss-20b/) | 53.5% OpenRouter Benchmarks | OpenAI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | gpt-oss-20b |
| [GPT-4.1](https://meirinho.com.br/en/ia/modelos/gpt-4-1/) | 52.3% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-4.1 |
| [Nemotron 3 Nano 30B A3B](https://meirinho.com.br/en/ia/modelos/nemotron-3-nano-30b-a3b/) | 52% OpenRouter Benchmarks | NVIDIA | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Nemotron 3 Nano 30B A3B |
| [GPT-5 Nano](https://meirinho.com.br/en/ia/modelos/gpt-5-nano/) | 48.3% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-5 Nano |
| [Qwen3-Coder](https://meirinho.com.br/en/ia/modelos/qwen3-coder/) | 48% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3-Coder |
| [Gemma 3 12B](https://meirinho.com.br/en/ia/modelos/gemma-3-12b-it/) | 48% OpenRouter Benchmarks | Google | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Gemma 3 12B |
| [Qwen3 VL 235B A22B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-vl-235b-a22b-instruct/) | 48% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 VL 235B A22B Instruct |
| [Qwen3-235B-A22B](https://meirinho.com.br/en/ia/modelos/qwen3-235b-a22b/) | 47.9% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3-235B-A22B |
| [Qwen3 Next 80B A3B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-next-80b-a3b-instruct/) | 47.5% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 Next 80B A3B Instruct |
| [Qwen3 30B A3B](https://meirinho.com.br/en/ia/modelos/qwen3-30b-a3b/) | 47% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 30B A3B |
| [Mistral Small 4](https://meirinho.com.br/en/ia/modelos/mistral-small-2603/) | 46% OpenRouter Benchmarks | Mistral AI | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Mistral Small 4 |
| [Qwen3 Coder 30B A3B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-coder-30b-a3b-instruct/) | 46% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 Coder 30B A3B Instruct |
| [GPT-4.1 Mini](https://meirinho.com.br/en/ia/modelos/gpt-4-1-mini/) | 45.1% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-4.1 Mini |
| [Qwen3 14B](https://meirinho.com.br/en/ia/modelos/qwen3-14b/) | 44% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 14B |
| [DeepSeek V3](https://meirinho.com.br/en/ia/modelos/deepseek-v3/) | 43.8% OpenRouter Benchmarks | DeepSeek | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | DeepSeek V3 |
| [Llama 4 Maverick](https://meirinho.com.br/en/ia/modelos/llama-4-maverick/) | 43.4% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Llama 4 Maverick |
| [GPT-4o](https://meirinho.com.br/en/ia/modelos/gpt-4o/) | 43.2% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-4o |
| [Qwen3 30B A3B Instruct 2507](https://meirinho.com.br/en/ia/modelos/qwen3-30b-a3b-instruct-2507/) | 41.5% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 30B A3B Instruct 2507 |
| [Qwen3 VL 30B A3B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-vl-30b-a3b-instruct/) | 41.3% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 VL 30B A3B Instruct |
| [Llama 3.3 70B Instruct](https://meirinho.com.br/en/ia/modelos/llama-3-3-70b-instruct/) | 39.9% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Llama 3.3 70B Instruct |
| [Qwen3 VL 8B Instruct](https://meirinho.com.br/en/ia/modelos/qwen3-vl-8b-instruct/) | 32.7% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen3 VL 8B Instruct |
| [GPT-4o-mini](https://meirinho.com.br/en/ia/modelos/gpt-4o-mini/) | 30.7% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-4o-mini |
| [Qwen2.5 72B Instruct](https://meirinho.com.br/en/ia/modelos/qwen-2-5-72b-instruct/) | 27.3% OpenRouter Benchmarks | Alibaba | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Qwen2.5 72B Instruct |
| [Llama 3.1 8B Instruct](https://meirinho.com.br/en/ia/modelos/llama-3-1-8b-instruct/) | 23.5% OpenRouter Benchmarks | Meta | open | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | Llama 3.1 8B Instruct |
| [GPT-4.1 Nano](https://meirinho.com.br/en/ia/modelos/gpt-4-1-nano/) | 14% OpenRouter Benchmarks | OpenAI | closed | [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline) | GPT-4.1 Nano |

## Model against model

- [GPT-6.1 Sol vs DeepSeek V4 Pro](https://meirinho.com.br/en/ia/modelos/comparar/deepseek-v4-pro-vs-gpt-6-1-sol/)
- [GPT-5 vs DeepSeek V3](https://meirinho.com.br/en/ia/modelos/comparar/deepseek-v3-vs-gpt-5/)
- [GPT-6.1 Sol vs Claude Opus 5.5](https://meirinho.com.br/en/ia/modelos/comparar/claude-opus-5-5-vs-gpt-6-1-sol/)
- [Claude Opus 5.5 vs Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/comparar/claude-opus-5-5-vs-gemini-4-argon/)
- [GPT-6.1 Sol vs Gemini 4 Argon](https://meirinho.com.br/en/ia/modelos/comparar/gemini-4-argon-vs-gpt-6-1-sol/)
- [DeepSeek V4 Pro vs Qwen3.8 Max](https://meirinho.com.br/en/ia/modelos/comparar/deepseek-v4-pro-vs-qwen3-8-max/)
- [GPT-6.1 Sol vs Qwen3.8 Max](https://meirinho.com.br/en/ia/modelos/comparar/gpt-6-1-sol-vs-qwen3-8-max/)
- [DeepSeek V4 Pro vs Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/comparar/deepseek-v4-pro-vs-kimi-k3-1-pro/)
- [DeepSeek V4 Pro vs GLM-5.3](https://meirinho.com.br/en/ia/modelos/comparar/deepseek-v4-pro-vs-glm-5-3/)
- [GPT-6.1 Sol vs Grok 4.7](https://meirinho.com.br/en/ia/modelos/comparar/gpt-6-1-sol-vs-grok-4-7/)
- [Gemini 3.8 Flash vs Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/comparar/claude-sonnet-5-5-vs-gemini-3-8-flash/)
- [Gemini 3.8 Flash vs DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/comparar/deepseek-v4-1-flash-vs-gemini-3-8-flash/)
- [Gemma 4 31B vs Qwen3.8 27B](https://meirinho.com.br/en/ia/modelos/comparar/gemma-4-31b-vs-qwen3-8-27b/)
- [gpt-oss-120b vs Qwen3.8 27B](https://meirinho.com.br/en/ia/modelos/comparar/gpt-oss-120b-vs-qwen3-8-27b/)
- [Claude Opus 5.5 vs Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/comparar/claude-opus-5-5-vs-claude-sonnet-5-5/)
- [Claude Opus 5.5 vs Kimi K3.1 Pro](https://meirinho.com.br/en/ia/modelos/comparar/claude-opus-5-5-vs-kimi-k3-1-pro/)
- [Qwen3-ASR vs Whisper large-v3](https://meirinho.com.br/en/ia/modelos/comparar/qwen3-asr-vs-whisper-large-v3/)
- [GLM-5.3 vs MiniMax M3](https://meirinho.com.br/en/ia/modelos/comparar/glm-5-3-vs-minimax-m3/)
- [GPT-6.2 Astra vs GPT-6 Astra](https://meirinho.com.br/en/ia/modelos/comparar/gpt-6-2-astra-vs-gpt-6-astra/)
- [ChatGPT](https://meirinho.com.br/en/ia/modelos/marca/chatgpt/)
- [Gemini](https://meirinho.com.br/en/ia/modelos/marca/gemini/)
- [Claude](https://meirinho.com.br/en/ia/modelos/marca/claude/)
- [DeepSeek](https://meirinho.com.br/en/ia/modelos/marca/deepseek/)
- [Grok](https://meirinho.com.br/en/ia/modelos/marca/grok/)
- [GPT-6.2 Astra vs Gemini 4 Carbon vs Claude Fable 5.2](https://meirinho.com.br/en/ia/modelos/comparar/claude-fable-5-2-vs-gemini-4-carbon-vs-gpt-6-2-astra/)
- [GPT-6.2 Astra vs Gemini 4 Carbon: benchmarks and pricing](https://meirinho.com.br/en/ia/modelos/comparar/gemini-4-carbon-vs-gpt-6-2-astra/)
- [GPT-6.2 Astra vs Claude Fable 5.2: benchmarks and pricing](https://meirinho.com.br/en/ia/modelos/comparar/claude-fable-5-2-vs-gpt-6-2-astra/)
- [Gemini 4 Carbon vs Claude Fable 5.2: benchmarks and pricing](https://meirinho.com.br/en/ia/modelos/comparar/claude-fable-5-2-vs-gemini-4-carbon/)
- [GPT-6.2 Astra vs Grok 4.7: benchmarks and pricing](https://meirinho.com.br/en/ia/modelos/comparar/gpt-6-2-astra-vs-grok-4-7/)
- [Gemini 4 Carbon vs Grok 4.7: benchmarks and pricing](https://meirinho.com.br/en/ia/modelos/comparar/gemini-4-carbon-vs-grok-4-7/)
- [Claude Fable 5.2 vs Grok 4.7: benchmarks and pricing](https://meirinho.com.br/en/ia/modelos/comparar/claude-fable-5-2-vs-grok-4-7/)
- [GPT-6.1 Sol vs Gemini 3.8 Flash vs Claude Sonnet 5.5](https://meirinho.com.br/en/ia/modelos/comparar/claude-sonnet-5-5-vs-gemini-3-8-flash-vs-gpt-6-1-sol/)
- [GPT-6 Luna vs Claude Haiku 5.5 vs DeepSeek V4.1 Flash](https://meirinho.com.br/en/ia/modelos/comparar/claude-haiku-5-5-vs-deepseek-v4-1-flash-vs-gpt-6-luna/)
- [ChatGPT vs Gemini: API models, tier by tier](https://meirinho.com.br/en/ia/modelos/comparar/chatgpt-vs-gemini/)
- [ChatGPT vs Claude: API models, tier by tier](https://meirinho.com.br/en/ia/modelos/comparar/chatgpt-vs-claude/)
- [Gemini vs Claude: API models, tier by tier](https://meirinho.com.br/en/ia/modelos/comparar/claude-vs-gemini/)
- [ChatGPT vs Grok: API models, tier by tier](https://meirinho.com.br/en/ia/modelos/comparar/chatgpt-vs-grok/)
- [ChatGPT vs DeepSeek: API models, tier by tier](https://meirinho.com.br/en/ia/modelos/comparar/chatgpt-vs-deepseek/)
- [ChatGPT vs Gemini vs Claude: API models compared](https://meirinho.com.br/en/ia/modelos/comparar/chatgpt-vs-claude-vs-gemini/)

## Sources

- [Epoch AI](https://epoch.ai/benchmarks), CC BY 4.0, latest version we checked.
- [OpenRouter Benchmarks](https://openrouter.ai/benchmarks), CC BY 4.0, latest version we checked.
- [OpenRouter](https://openrouter.ai/api/v1/models), dados públicos da API; exports de benchmarks e rankings em CC BY 4.0, latest version we checked.

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.

## How to cite

If you use this data, cite the page and the date you accessed it.

```
Adriano Meirinho. "Compare models". meirinho.com.br. https://meirinho.com.br/en/ia/modelos/comparar/. Accessed November 21, 2026.
```

Third-party data follows the license of each source. The full table is available as [CSV](https://meirinho.com.br/ia/modelos/dados.csv) and [JSON](https://meirinho.com.br/ia/modelos/dados.json).

[#](#)
