Mistral AI
Mistral: Mistral Medium 3.5
mistralai/mistral-medium-3-5
Mistral Medium 3.5 is a dense 128B instruction-following model from Mistral AI. It supports text and image inputs with text output, and is designed for agentic workflows, coding, and complex multi-step reasoning. It is particularly strong at reliable multi-tool calling and long-horizon tasks, with a 256K context window, configurable reasoning effort per request, and a custom vision encoder that handles variable image sizes and aspect ratios. Self-hostable on as few as four GPUs and available under open weights.
/ 1M tokens
/ 1M tokens
tokens
Limits
Context
262,144
tokens
Max output
209,715
80% of the context window
Model facts
Author
Mistral AI
Tokenizer
Mistral
Modality
Text + Image + File -> Text
Created
Apr 30, 2026
Reasoning
Publication status
Modalities
Generation controls
Capability scores
Better than 30% of models compared
Better than 44% of models compared
Better than 36% of models compared