# Gemma 3 12B: price, benchmarks, context

- **URL canônica:** https://meirinho.com.br/en/ia/modelos/gemma-3-12b-it/
- **Idioma:** en
- **Atualizado:** 2026-11-17

---
**insufficient data** — 3 of 4 benchmarks

- GPQA Diamond — low-resolution caveat (S index)
- GPQA Diamond (OpenRouter) — low-resolution caveat (S index)
- OTIS Mock AIME — saturated benchmark
- τ-bench Verified (companhia aérea) — low-resolution caveat (S index)

Methodology v2 · updated November 21, 2026 · [Technical tie](https://meirinho.com.br/en/ia/modelos/metodologia/#empate-tecnico) · [Minimum of 4 benchmarks](https://meirinho.com.br/en/ia/modelos/metodologia/#minimo-3-benchmarks) · [Methodology](https://meirinho.com.br/en/ia/modelos/metodologia/)

Google

# Gemma 3 12B

Benchmarks, price, context and versions

GPQA Diamond

Epoch AI

Input US$/1M

Third-party provider

Output US$/1M

Third-party provider

Context

tokens

## Direct answer

**Direct answer:** Gemma 3 12B is a model from Google. 39.5% on GPQA Diamond, per Epoch AI. On OpenRouter, US$ 0.05 per 1 million input tokens and US$ 0.15 per 1 million output tokens.

## ✓ Good for

- high-volume use (input price in the cheapest 25% on OpenRouter)
- running on your own infrastructure (open weights)

## ✕ Not the best for

- technical reasoning on science topics (164th of 195 on GPQA Diamond, per Epoch AI)
- technical reasoning as the model appears through the API (129th of 134 on GPQA Diamond (OpenRouter), per OpenRouter Benchmarks)

This block follows a rule of this section, using source data: if the model is in the top 25% of the metric, the item goes under "Good for"; if it is in the bottom half, it goes under "Not the best for". The middle band appears in neither. The price uses OpenRouter. Context uses OpenRouter. Scores use Epoch AI or OpenRouter Benchmarks, in the latest version we checked.

## Benchmarks

| GPQA Diamond[? what it is](https://meirinho.com.br/en/ia/modelos/glossario/gpqa-diamond/) | 39.5%**highest score on this metric 96.8%164º / 195 · [Epoch AI](https://epoch.ai/benchmarks/gpqa-diamond), latest version we checked |
| --- | --- |
| OTIS Mock AIME[?](https://meirinho.com.br/en/ia/modelos/glossario/otis-mock-aime/) | 16.7%**highest score on this metric 100%140º / 182 · [Epoch AI](https://epoch.ai/benchmarks/otis-mock-aime-2024-2025), latest version we checked |
| GPQA Diamond (OpenRouter)[?](https://meirinho.com.br/en/ia/modelos/glossario/gpqa-diamond-openrouter/) | 40.6%**highest score on this metric 96.5%129º / 134 · [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/gpqa-diamond), latest version we checked |
| τ-bench Verified (airline)[?](https://meirinho.com.br/en/ia/modelos/glossario/tau-bench/) | 48%**highest score on this metric 82%111º / 130 · [OpenRouter Benchmarks](https://openrouter.ai/benchmarks/tau2-bench-airline), latest version we checked |

## Input US$/1M · Output US$/1M

Input US$/1M

US$ 0.05

Output US$/1M

US$ 0.15

## Key specs

| Lab | [Google](https://meirinho.com.br/en/ia/modelos/lab/google/) |
| --- | --- |
| Origin | the US |
| Family | Gemma provisional data |
| Weights[?](https://meirinho.com.br/en/ia/modelos/glossario/pesos-abertos/) | open |
| License | no data |
| Also known as | no data |
| Context[?](https://meirinho.com.br/en/ia/modelos/glossario/janela-de-contexto/) | 131,072 tokens |
| Parameters | no data |
| Input | image, text |
| OpenRouter usage[?](https://meirinho.com.br/en/ia/modelos/glossario/uso-no-openrouter/) | 0.01% OpenRouter |
| Downloads (30 days)[?](https://meirinho.com.br/en/ia/modelos/glossario/downloads-hugging-face/) | no data |
| Use cases provisional data | [Text](https://meirinho.com.br/en/ia/modelos/uso/texto/) [Agents](https://meirinho.com.br/en/ia/modelos/uso/agentes/) |
| Dates in sources | no official dateMarch 12, 2025, published, per Epoch AIMarch 13, 2025, listed on OpenRouter |

Collected November 17, 2026

## Version history

- May 23, 2026 [gemma-4-12B-it](https://huggingface.co/google/gemma-4-12B-it) Hugging Face Hub, Hugging Face repo created
- May 23, 2026 [gemma-4-12B](https://huggingface.co/google/gemma-4-12B) Hugging Face Hub, Hugging Face repo created
- April 3, 2026 [Google: Gemma 4 26B A4B](https://openrouter.ai/google/gemma-4-26b-a4b-it) OpenRouter, listed on OpenRouter
- April 2, 2026 [Google: Gemma 4 31B](https://openrouter.ai/google/gemma-4-31b-it) OpenRouter, listed on OpenRouter
- March 12, 2026 [gemma-4-31B](https://huggingface.co/google/gemma-4-31B) Hugging Face Hub, Hugging Face repo created
- March 12, 2026 [gemma-4-26B-A4B](https://huggingface.co/google/gemma-4-26B-A4B) Hugging Face Hub, Hugging Face repo created
- March 11, 2026 [gemma-4-26B-A4B-it](https://huggingface.co/google/gemma-4-26B-A4B-it) Hugging Face Hub, Hugging Face repo created

Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch.

## Sources for this model

- Official model page, [Google](https://aistudio.google.com/docs/models).
- Third-party provider pricing, [OpenRouter](https://openrouter.ai/google/gemma-3-12b-it).
- No official launch announcement.
- Benchmarks: each score carries its own source, in the table above.

## FAQ

Gemma 3 12B is a model from Google, with open weights, used for text, agents.

On OpenRouter, US$ 0.05 per 1 million input tokens and US$ 0.15 per 1 million output tokens.

131,072 tokens, per OpenRouter.

Yes, it has open weights. License: no data.

39.5% on GPQA Diamond, per Epoch AI.

No data: there is no SWE-bench Verified measurement for it in open sources.

It was made by Google, a lab from the US.

provisional data

## Compare with

### Models from Google

- [Gemma 2 27B](https://meirinho.com.br/en/ia/modelos/gemma-2-27b-it/)
- [Gemma 2 9B](https://meirinho.com.br/en/ia/modelos/gemma-2-9b/)
- [Gemma 3 1B](https://meirinho.com.br/en/ia/modelos/gemma-3-1b/)
- [Gemma 3 27B](https://meirinho.com.br/en/ia/modelos/gemma-3-27b-it/)
- [Gemma 3 4B](https://meirinho.com.br/en/ia/modelos/gemma-3-4b-it/)
- [Gemma 4 26B A4B](https://meirinho.com.br/en/ia/modelos/gemma-4-26b-a4b-it/)
- [Gemma 4 31B](https://meirinho.com.br/en/ia/modelos/gemma-4-31b/)

[See lab](https://meirinho.com.br/en/ia/modelos/lab/google/)

## How to cite

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

```
Adriano Meirinho. "Gemma 3 12B". meirinho.com.br. https://meirinho.com.br/en/ia/modelos/gemma-3-12b-it/. 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).

[#](#)
