Mixtral: pricing, context window and benchmarks
Mixtral is a multimodal model from Mistral AI. It offers a 65,536-token context window, costs $3.48 per million input tokens and $12.73 per million output tokens, and averages 73.7 across the index's benchmark suite.
Pricing
Mixtral costs $3.48 per million input tokens and $12.73 per million output tokens. A blended 3:1 input-to-output mix works out to about $5.79 per million tokens. Cached input is listed at $0.43 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.48 | per 1M tokens |
| Output | $12.73 | per 1M tokens |
| Cached input | $0.43 | per 1M tokens |
| Blended 3:1 | $5.79 | estimate per 1M tokens |
| 1M in + 1M out | $16.21 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 77.3 | 269 | 55th | |
| GPQA Diamond | 71.3 | 234 | 61th | |
| SWE-bench Verified | 60.3 | 321 | 47th | |
| HumanEval | 75.9 | 330 | 45th | |
| MATH-500 | 78.5 | 249 | 59th | |
| AIME 2025 | 72.2 | 215 | 64th | |
| LiveCodeBench | 62.5 | 342 | 43th | |
| IFEval | 81.2 | 298 | 50th | |
| TruthfulQA | 76.4 | 297 | 51th | |
| HellaSwag | 82.8 | 326 | 46th | |
| MMMU | 72.7 | 228 | 62th | |
| Arena Elo | 1363.2 | 337 | 44th |
What Mixtral is good for
Mixtral is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.48 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $3.48 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
Trade-offs
- No standout weakness in the indexed specifications, but real-world behaviour still needs testing.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Grok 3 Mini | xAI | 200000 | $9.67 | 73.9 |
| Mistral Small 3 Chat | Mistral AI | 200000 | $3.54 | 73.7 |
| Mixtral | Mistral AI | 65536 | $3.48 | 73.7 |
| Gemma 3 | Google DeepMind | 65536 | $3.68 | 73.6 |
| GPT-5.2 Nano | OpenAI | 128000 | $9.73 | 73.5 |
| Yi-Large | 01.AI | 32768 | $3.40 | 73.5 |
FAQ
How much does Mixtral cost?
Mixtral is listed at $3.48 per million input tokens and $12.73 per million output tokens. Cached input is $0.43 per million tokens. A blended 3:1 input-to-output mix is about $5.79 per million tokens.
What is the context window of Mixtral?
Mixtral has a context window of 65,536 tokens and can emit up to 16,384 output tokens per request in this index.
Is Mixtral open weights?
Yes. Mixtral is listed as an open-weight model under a open weights licence.
How does Mixtral score on benchmarks?
Mixtral averages 73.7 across the scored benchmarks in this index, ranking #283 of 600 models. Its strongest relative result is Arena Elo at 1363.2.
What is Mixtral best for?
Mixtral is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.48 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.