Mistral Medium 3.5
Mistral AI llmDense open-weight frontier-class model optimised for agentic and coding use, previously an API-only tier.
overall ⌄
56.7
43 of 125 ranked
price · $/M tokens
1.5 / 7.5
in / out, as of 2026-07-25
to run it yourself ⌄
77 GB
minimum · 307 GB recommended
Benchmarks
What its maker published, and what anyone else measured. Every figure links the document it came from.
Cheaper, and at least as good
On a 1,000-in, 1,000-out call these cost less and rank no lower — a hard comparison to argue with, and a narrow one. Value covers what it misses.
| DeepSeek V4-Flash DeepSeek | $0.00042 | overall 57.9 |
| Claude Haiku 5.5 Anthropic | $0.00060 | overall 65.3 |
| Qwen3.8-Flash-Next Alibaba | $0.00062 | overall 61.2 |
| DeepSeek V4-Pro DeepSeek | $0.00130 | overall 65.6 |
| DeepSeek-V4.1-Flash DeepSeek | $0.00150 | overall 59.2 |
Lineage
No recorded lineage — root or standalone model.
Metadata
| org | Mistral AI |
| released | 2026-04-28 |
| params | 128B |
| license | modified-mit |
| hf id | mistralai/Mistral-Medium-3.5-128B |