Mistral: Mistral Medium 3.5

Image input Text input Text output Unavailable
Author's Description

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...

Key Specifications
Cost
$$$$
Context
262K
Parameters
128B (Rumoured)
Released
Apr 30, 2026
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Tools Response Format Stop Frequency Penalty Tool Choice Include Reasoning Temperature Max Tokens Reasoning Structured Outputs Presence Penalty Top P Seed
Features

This model supports the following features:

Structured Outputs Tools Reasoning Response Format
Performance Summary

Mistral Medium 3.5, a dense 128B instruction-following model from Mistral AI, demonstrates strong overall performance, particularly in speed and reliability. It consistently ranks among the fastest models, achieving the 92nd percentile across eight benchmarks. Its reliability is exceptional, boasting a 100% success rate across all evaluated benchmarks, indicating minimal technical failures. While its pricing is moderate, ranking in the 31st percentile, its performance often justifies the cost. The model exhibits key strengths in hallucination prevention, achieving perfect 100% accuracy, making it the most accurate model at its price point and among models of comparable speed. It also excels in coding, with 95.0% accuracy (92nd percentile), and mathematics, scoring 94.0% (80th percentile). General knowledge and ethics also show strong results at 99.5% and 99.0% accuracy, respectively. Instruction following and reasoning capabilities are solid, with 71.0% and 80.0% accuracy. Email classification is competent at 98.0%. Overall, Mistral Medium 3.5 is a highly reliable and fast model, particularly well-suited for tasks requiring precision in coding, mathematical problem-solving, and avoiding factual inaccuracies.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $1.5
Completion $7.5

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Mistral
Mistral | mistralai/mistral-medium-3.5-20260430 262K $1.5 / 1M tokens $7.5 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by mistralai