Mistral Models
Mistral logo

Devstral Medium

by Mistral

Devstral Medium is a high-performance code generation and agentic reasoning model developed jointly by Mistral AI and All Hands AI. Positioned as a step up from Devstral Small, it achieves 61.6% on SWE-Bench Verified, placing it ahead of Gemini 2.5 Pro and GPT-4.1 in code-related tasks, at a fraction of the cost. It is designed for generalization across prompt styles and tool use in code agents and frameworks. Devstral Medium is available via API only (not open-weight), and supports enterprise deployment on private infrastructure, with optional fine-tuning capabilities.

Chat with Devstral Medium
Input Price$0.40/1M tokens
Output Price$2.00/1M tokens
Intelligence18.6
Coding15.9

Specifications

Technical details and pricing.

ProviderMistral
Context Window131,072 tokens
Release DateJul 10, 2025
ModalitiesText

Benchmarks

12 benchmark scores from Artificial Analysis.

GPQA49.2%
MMLU Pro70.8%
HLE3.8%
LiveCodeBench33.7%
MATH 50070.7%
AIME 20254.7%
AIME6.7%
SciCode29.4%
LCR28.7%
IFBench29.9%
Tau219.9%
TerminalBench Hard9.1%

Composite Indices

Intelligence, Coding, Math

Standard Benchmarks

Academic and industry benchmarks

Frequently Asked Questions

What is Devstral Medium good for?

Use Devstral Medium for everyday tasks like writing, summarizing, brainstorming, and getting clear explanations.

How much does Devstral Medium cost?

Pricing is based on usage. Current rates are $0.40/1M tokens for input and $2.00/1M tokens for output.

Can I try Devstral Medium for free?

Yes. You can start a chat instantly and test the model before deciding on a plan.

Does Devstral Medium support images or audio?

Devstral Medium focuses on text-based tasks.

Benchmarks and pricing are sourced from Artificial Analysis where available. OpenRouter specs are used as a fallback.