GPQA Diamond
?what it ishighest score on this metric 96.4%
44º / 198 · Epoch AI, latest version we checked
Mistral Large 4
Benchmarks, price, context and versions
Also searched as
Mistral Large, Mistral AI, mistral large 4 benchmark results, mistral large 4 benchmark, mistral large 4 huggingface
Direct answer: Mistral Large 4 is a model from Mistral AI. 90.2% on GPQA Diamond, per Epoch AI. On OpenRouter, US$ 0.70 per 1 million input tokens and US$ 2.10 per 1 million output tokens.
✓ Good for
- technical reasoning on science topics (#44 of 198 on GPQA Diamond)
- calculation and problems with an exact answer (#18 of 186 on OTIS Mock AIME)
- very long documents or code (a window of 1 million tokens or more, according to OpenRouter)
- running on your own infrastructure (open weights)
✕ Not the best for
- the ceiling of reasoning on research-level math (#44 of 86 on FrontierMath (tiers 1 to 3))
- answering short facts with little invention (#73 of 87 on SimpleQA Verified)
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. Price and context use OpenRouter; scores use Epoch AI or OpenRouter Benchmarks, in the latest version we checked.
Benchmarks
Key specs
- Lab
- Mistral AI
- Origin
- France
- Family
- Mistral Large provisional data
- Weights?
- open
- License
- no data
- Input US$/1M?
- US$ 0.70
- Output US$/1M
- US$ 2.10
- Context?
- 1,048,576 tokens
- Parameters
- no data
- Input
- image, text
- OpenRouter usage?
- no data
- Downloads (30 days)?
- no data
- Dates in sources
- 2026, published, per Epoch AI
- 2026, listed on OpenRouter
Version history
- Mistral: Mistral Large 4 OpenRouter, listed on OpenRouter
- Mistral-Large-4-1T-A52B Hugging Face Hub, Hugging Face repo created
- Mistral: Mistral Large 3 2512 OpenRouter, listed on OpenRouter
- Mistral-Large-3-675B-Base-2512 Hugging Face Hub, Hugging Face repo created
- Mistral-Large-3-675B-Instruct-2512 Hugging Face Hub, Hugging Face repo created
- Mistral-Large-3-675B-Instruct-2512-Eagle Hugging Face Hub, Hugging Face repo created
- Mistral-Large-3-675B-Instruct-2512-BF16 Hugging Face Hub, Hugging Face repo created
- Mistral Large 2407 OpenRouter, listed on OpenRouter
- Mistral-Large-Instruct-2411 Hugging Face Hub, Hugging Face repo created
- Mistral-Large-Instruct-2407 Hugging Face Hub, Hugging Face repo created
- Mistral Large OpenRouter, listed on OpenRouter
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch.
FAQ
How much does Mistral Large 4 cost?
On OpenRouter, US$ 0.70 per 1 million input tokens and US$ 2.10 per 1 million output tokens.
Is Mistral Large 4 open source?
Yes, it has open weights. License: no data.
What is Mistral Large 4's GPQA Diamond score?
90.2% on GPQA Diamond, per Epoch AI.
What is Mistral Large 4?
Mistral Large 4 is a model from Mistral AI, with open weights, used for text, agents.
What is the context window of Mistral Large 4?
1,048,576 tokens, per OpenRouter.
Is Mistral Large 4 good for coding?
No data: there is no SWE-bench Verified measurement for it in open sources.
Who made Mistral Large 4?
It was made by Mistral AI, a lab from France.
provisional data Questions ordered by what shows up in public Google search suggestions (no volume). Answers use source data only.
Models from Mistral AI
Sources
- Epoch AI, 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.