Mistral AI: model lineup and pricing overview
Mistral AI model lineup with prices, context windows and average benchmark scores. European model lab focused on open and commercial LLMs.
What the data shows
Mistral AI has 35 models in the index. The average input price is $2.71 and the average output price is $12.15 per million tokens.
The strongest model by average score is mistral-large-3. The cheapest by input price is Mixtral 1.5B.
How to read the lineup
- Start with the best model for quality-sensitive work.
- Use the cheapest model for high-volume tasks.
- Check open-weight availability if self-hosting matters.
- Keep a second provider for fallback.
Mistral AI models
| # | Model | Input / 1M | Output / 1M | Context | Avg score |
|---|---|---|---|---|---|
| 1 | Mistral Large 3 | $5.41 | $23.83 | 128,000 | 81.7 |
| 2 | Mistral Large 3 Instruct | $5.31 | $22.47 | 400,000 | 81.1 |
| 3 | Magistral | $4.75 | $21.16 | 200,000 | 79.8 |
| 4 | Mistral Large 3 Chat | $4.90 | $20.04 | 1,000,000 | 79.1 |
| 5 | Magistral Instruct | $4.48 | $22.65 | 65,536 | 78.4 |
| 6 | Mistral Large 3 Base | $4.47 | $23.57 | 65,536 | 78.2 |
| 7 | Codestral | $4.19 | $16.36 | 65,536 | 77.6 |
| 8 | Magistral Chat | $4.33 | $16.06 | 128,000 | 76.8 |
| 9 | Codestral Chat | $3.89 | $18.19 | 32,768 | 75.8 |
| 10 | Magistral Base | $3.86 | $17.23 | 65,536 | 75.6 |
| 11 | Mistral Small 3 | $3.90 | $20.43 | 32,768 | 75.3 |
| 12 | Codestral Instruct | $4.01 | $16.29 | 65,536 | 75.1 |
| 13 | Mistral Small 3 Instruct | $3.97 | $20.95 | 32,768 | 75.1 |
| 14 | Mistral Small 3 Chat | $3.54 | $16.77 | 200,000 | 73.7 |
| 15 | Mixtral | $3.48 | $12.73 | 65,536 | 73.7 |
| 16 | Codestral Base | $3.31 | $16.95 | 8,192 | 73.1 |
| 17 | Mixtral Instruct | $3.35 | $18.19 | 65,536 | 72.9 |
| 18 | Mistral Small 3 Base | $3.19 | $11.34 | 16,384 | 72.7 |
| 19 | Mixtral Chat | $3.30 | $13.04 | 65,536 | 72.6 |
| 20 | Mixtral Base | $2.95 | $12.22 | 65,536 | 71.1 |
Methodology
This story uses the AI & LLM Index dataset: 600 models, 40 providers and 12 benchmarks. Prices are listed per million tokens. Scores are modelled estimates on a 0–100 scale and are not official benchmark results. Figures are rounded for readability.
Related
FAQ
What does the Mistral AI: model lineup and pricing overview data show?
Mistral AI model lineup with prices, context windows and average benchmark scores. European model lab focused on open and commercial LLMs.
Are the numbers official?
No. Prices and scores are compiled for this demo index. Benchmark values are modelled estimates, not official results. Verify figures with the provider or benchmark owner before making decisions.
How should I use this ranking?
Use it to form a shortlist. Check the specific model pages, compare price and context, and test the shortlist on your own examples before committing.