Magistral Base: pricing, context window and benchmarks
Magistral Base is a multimodal model from Mistral AI. It offers a 65,536-token context window, costs $3.86 per million input tokens and $17.23 per million output tokens, and averages 75.6 across the index's benchmark suite.
Pricing
Magistral Base costs $3.86 per million input tokens and $17.23 per million output tokens. A blended 3:1 input-to-output mix works out to about $7.20 per million tokens. Cached input is listed at $0.55 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.86 | per 1M tokens |
| Output | $17.23 | per 1M tokens |
| Cached input | $0.55 | per 1M tokens |
| Blended 3:1 | $7.20 | estimate per 1M tokens |
| 1M in + 1M out | $21.09 | 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 | 80.7 | 159 | 74th | |
| GPQA Diamond | 73.5 | 185 | 69th | |
| SWE-bench Verified | 65.6 | 201 | 67th | |
| HumanEval | 78.5 | 281 | 53th | |
| MATH-500 | 76.3 | 293 | 51th | |
| AIME 2025 | 69.1 | 279 | 54th | |
| LiveCodeBench | 69.7 | 198 | 67th | |
| IFEval | 83.5 | 230 | 62th | |
| TruthfulQA | 77.6 | 266 | 56th | |
| HellaSwag | 85.0 | 245 | 59th | |
| MMMU | 72.2 | 236 | 61th | |
| Arena Elo | 1378.9 | 276 | 54th |
What Magistral Base is good for
Magistral Base 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 $3.86 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 $3.86 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 |
|---|---|---|---|---|
| Llama 3 Lambda | Lambda | 65536 | $3.90 | 75.6 |
| Llama 3 Novita Instruct | Novita AI | 32768 | $3.94 | 75.6 |
| Magistral Base | Mistral AI | 65536 | $3.86 | 75.6 |
| Reka Flash Reasoning | Reka AI | 32768 | $10.70 | 75.6 |
| Llama 3 DeepInfra Instruct | DeepInfra | 65536 | $3.84 | 75.5 |
| Yi-Lightning Chat | 01.AI | 32768 | $3.79 | 75.5 |
FAQ
How much does Magistral Base cost?
Magistral Base is listed at $3.86 per million input tokens and $17.23 per million output tokens. Cached input is $0.55 per million tokens. A blended 3:1 input-to-output mix is about $7.20 per million tokens.
What is the context window of Magistral Base?
Magistral Base has a context window of 65,536 tokens and can emit up to 16,384 output tokens per request in this index.
Is Magistral Base open weights?
Yes. Magistral Base is listed as an open-weight model under a open weights licence.
How does Magistral Base score on benchmarks?
Magistral Base averages 75.6 across the scored benchmarks in this index, ranking #236 of 600 models. Its strongest relative result is Arena Elo at 1378.9.
What is Magistral Base best for?
Magistral Base 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 $3.86 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.