Mistral: Mistral Medium 3.5

Mistral Medium 3.5 is a mid-tier model from Mistral AI that accepts text, image, and file inputs. It offers a 262,144-token context window and can return up to 209,715 tokens in a single completion, which suits long documents and lengthy generated outputs. It supports tool calling and a reasoning mode. The listing does not confirm structured output support or open weights, so teams that depend on either should verify before committing. Pricing is $1.50 per million input tokens and $7.50 per million output tokens. Its blended benchmark score of 53.0 comes from only three independent benchmarks, so treat its standing as provisional. Within that set, it scores 77.4 on the AI Index coding measure and 31.3 on the agentic measure, against an overall index of 49.4. That pattern makes it worth shortlisting for coding and long-context work with mixed inputs. Buyers who need autonomous multi-step agents should test it against stronger agentic performers first.