mistralai

Mistral: Mistral Medium 3.5

Mistral Medium 3.5, developed by MistralAI, boasts an extensive context length of 262,144 tokens and supports multiple input modalities including text, images, and files. It also offers reasoning capabilities and can handle tools, making it versatile for a range of tasks. This model shines in specific areas like coding but has lower performance in agentic tasks. With its blended benchmark score of 52.5 across three independent benchmarks, Mistral Medium 3.5 is best suited for those requiring robust text processing and integrated tool support at a reasonable cost, despite its higher pricing for input and output tokens.

Quality Score
100/100
price + capability + benchmarks
Input Price
$1.50
per 1M tokens
Output Price
$7.50
per 1M tokens
Context Window
262,144
tokens

Benchmark results

Independent, published benchmarks. Blended score 52.5 across 3 benchmarks, last refreshed 2026-08-03. How scoring works →

BenchmarkMeasuresScore
AI Index broad capability composite 49.4
AI Index Coding software engineering tasks 77.4
AI Index Agentic multi-step tool-using tasks 31.3
Model ID
mistralai/mistral-medium-3-5
Vendor
mistralai
Released
April 2026
Tokenizer
Mistral
Input Modalities
text, image, file
Output Modalities
text
Max Output
default
Tool Calling
✓ supported
Structured Output
✓ supported
Reasoning Mode
✓ supported
Vision
✓ accepts images
Audio
no
Moderated
no

What it costs in practice

Computed from the current $1.50/M input and $7.50/M output rates. Run your own numbers →

JobTokensCost
Summarize a 50-page report 30k in / 1.5k out $0.06
Classify 1,000 customer emails 500k in / 50k out $1.12
A month of a busy support chatbot 5M in / 2M out $22.50

Similar models

Quick answers

How much does Mistral: Mistral Medium 3.5 cost?
$1.50 per million input tokens and $7.50 per million output tokens.
What is Mistral: Mistral Medium 3.5's context window?
262,144 tokens, roughly 393 pages of text in a single request.
Does Mistral: Mistral Medium 3.5 support tool calling?
It supports tool calling, structured output, a reasoning mode.
Can Mistral: Mistral Medium 3.5 process images?
Yes, it accepts image input.