Mistral Large 3
Mistral's largest general model for enterprise agents, coding, and multilingual reasoning
- Strongest at
- Writing Better than 10% of the past year's models · one benchmark so far
- Price per 1M tokens
- $0.5 / $1.5 input / output · Mistral
- Context window
- 262.1K tokens in · 262.1K out
- Since the last one
- 379 days after Mistral Large 2.1
Capability profile
Scored against 91 models released in the past year. 50 is the average model; further out is better.
No results yet: Coding, Reasoning, Math, Science, Computer use
Where these scores come from
Each capability combines the published benchmarks listed here, rescaled so 50 is the reference average and 15 points is one standard deviation. A hollow dot rests on a single benchmark.
- Writing Lech Mazur Writing comparison
- Reading & documents GDP.pdf (Surge AI)
- Visual reasoning GDP.pdf (Surge AI)
Price against capability
The shaded corner holds anything better and cheaper. Prices blend input and output 3 to 1.
13 models from the past year score higher on Writing for less: Muse Glimmer 30B, DeepSeek V4 Flash Vision Exp, DeepSeek V4.1 Flash and 10 more.
14 models from the past year score higher on Reading & documents for less: Qwen3.5 9B, DeepSeek V4.1 Flash, DeepSeek V4 Flash 0731 and 11 more.
10 models from the past year score higher on Visual reasoning for less: Claude Haiku 5.5, GPT-6 Luna, DeepSeek V4.1 Flash and 7 more.
Mistral $0.5 / $1.5 · OpenRouter $0.5 / $1.5 · Cached input $0.05
Since Mistral Large 2.1
5 releases in the Mistral Large line.
A newer release is out: Mistral Large 4.
Mistral's past year
15 release days in the past year, about one every 24 days. The next is expected around Nov 16, 2026. Forecasts ↗
Specifications
Specifications and prices from Models.dev; availability and limits may differ by provider. Benchmarks from Epoch AI (CC BY 4.0), LiveBench and the other sources named above.