Mistral Large 3 Base: pricing, context window and benchmarks
Mistral Large 3 Base is a multimodal model from Mistral AI. It offers a 65,536-token context window, costs $4.47 per million input tokens and $23.57 per million output tokens, and averages 78.2 across the index's benchmark suite.
Pricing
Mistral Large 3 Base costs $4.47 per million input tokens and $23.57 per million output tokens. A blended 3:1 input-to-output mix works out to about $9.25 per million tokens. Cached input is listed at $0.54 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $4.47 | per 1M tokens |
| Output | $23.57 | per 1M tokens |
| Cached input | $0.54 | per 1M tokens |
| Blended 3:1 | $9.25 | estimate per 1M tokens |
| 1M in + 1M out | $28.04 | 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 | 78.9 | 214 | 64th | |
| GPQA Diamond | 75.5 | 132 | 78th | |
| SWE-bench Verified | 64.3 | 227 | 62th | |
| HumanEval | 85.0 | 118 | 80th | |
| MATH-500 | 84.5 | 116 | 81th | |
| AIME 2025 | 70.8 | 241 | 60th | |
| LiveCodeBench | 72.3 | 143 | 76th | |
| IFEval | 83.3 | 237 | 61th | |
| TruthfulQA | 81.0 | 157 | 74th | |
| HellaSwag | 88.3 | 131 | 78th | |
| MMMU | 75.8 | 142 | 76th | |
| Arena Elo | 1420.3 | 148 | 75th |
What Mistral Large 3 Base is good for
Mistral Large 3 Base is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $4.47 per million input tokens, it sits in the mid-range of the index, and the 65,536-token context window is enough for large files and multi-file diffs.
Strengths
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
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 |
|---|---|---|---|---|
| GLM-4.5 | Zhipu AI | 200000 | $4.55 | 78.2 |
| Hunyuan-Large | Tencent | 65536 | $4.33 | 78.2 |
| Hunyuan-Large Instruct | Tencent | 65536 | $4.39 | 78.2 |
| Llama 3.3 | Meta AI | 128000 | $4.56 | 78.2 |
| Mistral Large 3 Base | Mistral AI | 65536 | $4.47 | 78.2 |
| QwQ Chat | Alibaba Cloud | 65536 | $4.28 | 78.2 |
FAQ
How much does Mistral Large 3 Base cost?
Mistral Large 3 Base is listed at $4.47 per million input tokens and $23.57 per million output tokens. Cached input is $0.54 per million tokens. A blended 3:1 input-to-output mix is about $9.25 per million tokens.
What is the context window of Mistral Large 3 Base?
Mistral Large 3 Base has a context window of 65,536 tokens and can emit up to 65,536 output tokens per request in this index.
Is Mistral Large 3 Base open weights?
Yes. Mistral Large 3 Base is listed as an open-weight model under a open weights licence.
How does Mistral Large 3 Base score on benchmarks?
Mistral Large 3 Base averages 78.2 across the scored benchmarks in this index, ranking #167 of 600 models. Its strongest relative result is Arena Elo at 1420.3.
What is Mistral Large 3 Base best for?
Mistral Large 3 Base is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $4.47 per million input tokens, it sits in the mid-range of the index, and the 65,536-token context window is enough for large files and multi-file diffs.