AI models

SimpleQA Verified leaderboard: AI models ranked

87 models ranked. The top score on SimpleQA Verified belongs to Claude Fable 5.2, at 94% per Epoch AI.

SimpleQA Verified, on the Epoch AI leaderboard (CC BY 4.0, latest version we checked), asks short factual questions with a single right answer, reviewed to remove answer-key errors. The score is the share of correct answers. Here are the 87 models in this section with a score on SimpleQA Verified, per Epoch AI. Rank is shown as a range, because scores within the margin are in a statistical tie.

SimpleQA Verified

94%

Claude Fable 5.2

models

87

01

Model%Source
Claude Fable 5.2Anthropic · closed94% 1st of 87 top scoreEpoch AI
GPT-6.2 AstraOpenAI · closed78.5% between 2nd and 3rd of 87 Epoch AI
Gemini 4 CarbonGoogle · closed77.7% between 2nd and 3rd of 87 Epoch AI
Gemini 4 ArgonGoogle · closed76% between 4th and 5th of 87 Epoch AI
GPT-6 AstraOpenAI · closed75.6% between 4th and 5th of 87 Epoch AI
GPT-6.1 SolOpenAI · closed73.9% between 6th and 7th of 87 Epoch AI
Gemini 3.1 ProGoogle · closed73.5% between 6th and 7th of 87 Epoch AI
Claude Opus 5.5Anthropic · closed72.2% 8th of 87 Epoch AI
Claude Fable 5.1Anthropic · closed70.8% between 9th and 10th of 87 Epoch AI
Claude Fable 5Anthropic · closed70.7% between 9th and 12th of 87 Epoch AI
Gemini 3.8 FlashGoogle · closed69.7% between 10th and 13th of 87 Epoch AI
GPT-5.6 SolOpenAI · closed69.7% between 10th and 13th of 87 Epoch AI
Gemini 3.7 FlashGoogle · closed69.2% between 11th and 13th of 87 Epoch AI
Gemini 3 FlashGoogle · closed66.8% between 14th and 16th of 87 Epoch AI
Gemini 3.5 FlashGoogle · closed66.2% between 14th and 16th of 87 Epoch AI
Gemini 3.6 FlashGoogle · closed66.2% between 14th and 16th of 87 Epoch AI
GPT-5.5OpenAI · closed63% between 17th and 18th of 87 Epoch AI
Kimi K3.1 ProMoonshot AI · open62.2% between 17th and 18th of 87 Epoch AI
GPT-6 SolOpenAI · closed60.7% between 19th and 21st of 87 Epoch AI
Muse Spark 1.2Meta · closed60.3% between 19th and 21st of 87 Epoch AI
Claude Opus 5Anthropic · closed59.9% between 19th and 21st of 87 Epoch AI
Muse Spark 1.1Meta · closed57.8% between 22nd and 23rd of 87 Epoch AI
GPT-6.1 TerraOpenAI · closed57.3% between 22nd and 24th of 87 Epoch AI
Muse Spark OpenMeta · open56.6% between 23rd and 26th of 87 Epoch AI
Grok 4.7xAI · closed56% between 24th and 26th of 87 Epoch AI
Qwen3.7 MaxAlibaba · closed55.8% between 24th and 26th of 87 Epoch AI
Claude Opus 4.8Anthropic · closed53% between 27th and 28th of 87 Epoch AI
Qwen 3.6 Max (Preview)Alibaba · undisclosed52% between 27th and 29th of 87 Epoch AI
Claude Opus 4.7Anthropic · closed51.7% between 28th and 29th of 87 Epoch AI
Kimi K3Moonshot AI · open50.6% between 30th and 31st of 87 Epoch AI
GPT-5OpenAI · closed50.1% between 30th and 33rd of 87 Epoch AI
o3OpenAI · closed49.4% between 31st and 34th of 87 Epoch AI
Grok 4.6xAI · closed49.3% between 31st and 35th of 87 Epoch AI
Qwen3-MaxAlibaba · closed48.7% between 32nd and 36th of 87 Epoch AI
Grok 4.5xAI · closed48.3% between 33rd and 37th of 87 Epoch AI
GPT-5.1OpenAI · closed48% between 34th and 39th of 87 Epoch AI
Qwen3.8 MaxAlibaba · closed47.3% between 35th and 41st of 87 Epoch AI
Claude Opus 4.6Anthropic · closed47% between 36th and 41st of 87 Epoch AI
DeepSeek V4 ProDeepSeek · open47% between 36th and 41st of 87 Epoch AI
Claude Sonnet 5.5Anthropic · closed46.5% between 37th and 42nd of 87 Epoch AI
GPT-5.4 ProOpenAI · closed46.3% between 37th and 42nd of 87 Epoch AI
Claude Opus 4.5Anthropic · closed45.7% between 40th and 44th of 87 Epoch AI
GPT-5.4OpenAI · closed45.1% between 42nd and 45th of 87 Epoch AI
Doubao Seed 2.1 ProByteDance · closed44.7% between 42nd and 45th of 87 Epoch AI
Qwen3.6 PlusAlibaba · closed44.1% between 43rd and 46th of 87 Epoch AI
GPT-5.6 TerraOpenAI · closed43.2% between 45th and 47th of 87 Epoch AI
DeepSeek V4.1 FlashDeepSeek · open42.9% between 46th and 48th of 87 Epoch AI
Doubao Seed Evolving (Nov 2026)ByteDance · closed41.9% between 47th and 52nd of 87 Epoch AI
GPT-6 LunaOpenAI · closed41.4% between 48th and 53rd of 87 Epoch AI
o1OpenAI · closed41.1% between 48th and 53rd of 87 Epoch AI
GLM-5.3Zhipu AI (Z.ai) · open41% between 48th and 53rd of 87 Epoch AI
GPT-5.6 LunaOpenAI · closed41% between 48th and 53rd of 87 Epoch AI
Qwen3-235B-A22BAlibaba · open40.4% between 49th and 54th of 87 Epoch AI
Doubao-Seed-2.0 ProByteDance · closed39.4% between 53rd and 54th of 87 Epoch AI
GPT-5.2OpenAI · closed37.1% between 55th and 56th of 87 Epoch AI
Kimi K2.7 CodeMoonshot AI · open36.5% between 55th and 57th of 87 Epoch AI
Claude Sonnet 4.6Anthropic · closed35.5% between 56th and 58th of 87 Epoch AI
Kimi K2.6Moonshot AI · open34.9% between 57th and 61st of 87 Epoch AI
Kimi K2.5Moonshot AI · open34.3% between 58th and 63rd of 87 Epoch AI
GLM-5.2Zhipu AI (Z.ai) · open34.2% between 58th and 64th of 87 Epoch AI
GLM 5.1Zhipu AI (Z.ai) · open34% between 58th and 64th of 87 Epoch AI
Claude Sonnet 5Anthropic · closed33.7% between 59th and 64th of 87 Epoch AI
DeepSeek V4 Flash 0731DeepSeek · open33.6% between 59th and 64th of 87 Epoch AI
Grok 4.3 BetaxAI · closed33.2% between 60th and 65th of 87 Epoch AI
GLM 4.7Zhipu AI (Z.ai) · open32.2% between 64th and 66th of 87 Epoch AI
Doubao Seed 2.0 Lite (Apr 2026)ByteDance · closed31.5% between 65th and 68th of 87 Epoch AI
GPT-4.1OpenAI · closed31.1% between 66th and 69th of 87 Epoch AI
Claude Sonnet 4.5Anthropic · closed30.7% between 66th and 70th of 87 Epoch AI
Grok 4.20xAI · closed30.2% between 67th and 71st of 87 Epoch AI
Doubao Seed 2.1 TurboByteDance · closed29.7% between 68th and 71st of 87 Epoch AI
GPT-5.4 MiniOpenAI · closed29.4% between 69th and 71st of 87 Epoch AI
Qwen3.5 Plus 2026-02-15Alibaba · undisclosed25.4% 72nd of 87 Epoch AI
Claude Haiku 5.5Anthropic · closed23.8% 73rd of 87 Epoch AI
Mistral Large 4Mistral AI · open22.4% between 74th and 75th of 87 Epoch AI
GPT-5 MiniOpenAI · closed21.6% between 74th and 75th of 87 Epoch AI
Qwen3.5-FlashAlibaba · undisclosed20.3% between 76th and 77th of 87 Epoch AI
o4-miniOpenAI · closed19.6% between 76th and 77th of 87 Epoch AI
Doubao-Seed-1.6ByteDance · closed17.4% 78th of 87 Epoch AI
Qwen3.6 FlashAlibaba · undisclosed15.9% between 79th and 80th of 87 Epoch AI
o3-miniOpenAI · closed15.3% between 79th and 80th of 87 Epoch AI
Claude Haiku 4.5Anthropic · closed13.2% between 81st and 83rd of 87 Epoch AI
GPT-4.1 MiniOpenAI · closed12.7% between 81st and 85th of 87 Epoch AI
Claude 3 OpusAnthropic · closed12.6% between 81st and 85th of 87 Epoch AI
GPT-5 NanoOpenAI · closed11.7% between 82nd and 85th of 87 Epoch AI
GPT-5.4 NanoOpenAI · closed11.7% between 82nd and 85th of 87 Epoch AI
GPT-4o-miniOpenAI · closed8.3% 86th of 87 Epoch AI
GPT-4.1 NanoOpenAI · closed6% 87th of 87 Epoch AI

FAQ

Which model has the top score on SimpleQA Verified?

The top score on SimpleQA Verified belongs to Claude Fable 5.2, at 94% per Epoch AI.

What does SimpleQA Verified measure?

SimpleQA Verified, on the Epoch AI leaderboard (CC BY 4.0, latest version we checked), asks short factual questions with a single right answer, reviewed to remove answer-key errors. The score is the share of correct answers.

What does “statistical tie with the top score” mean?

It means the gap to the top score is within the 95% margin, which is never below 1 point. On SimpleQA Verified, near 90%, that margin is about 2.6 points. That is why rank is shown as a range, «between Xth and Yth of N»: among tied models, the order may be measurement noise.

Which models are in this ranking?

The 87 models in this section with a published score on SimpleQA Verified, per Epoch AI. No version inherits another's score.

How to read

  1. Benchmark scores are percentages, and higher is better.
  2. Rank is shown as a range, «1st of 87». Within the range, the gap to neighboring models falls within the benchmark's margin. The «top score» badge marks the model with the benchmark's top score, and the «statistical tie with the top score» badge marks models whose gap to it is within the margin.
  3. On a saturated benchmark, the table shows the scores with the badge and without a rank order.

Sources

  • Epoch AI, CC BY 4.0, latest version we checked.
  • OpenRouter Benchmarks, CC BY 4.0, latest version we checked.
  • OpenRouter, 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. "SimpleQA Verified leaderboard: AI models ranked". meirinho.com.br. https://meirinho.com.br/en/ia/modelos/benchmark/simpleqa-verified/. Accessed November 21, 2026.

Third-party data follows the license of each source. The full table is available as CSV and JSON.

Markdown