devstral-medium

completions

On:mistralaiopenrouter

devstral-medium

completions

Published by: mistralaiProviders:

mistralai openrouter

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.

Provider	Input	Output
mistralai	$0.4 / 1M tokens	$2 / 1M tokens
openrouter	$0.4 / 1M tokens	$2 / 1M tokens

Released

Jul 10, 2025

Knowledge

Jan 11, 2025

Context

128K

Input

$0.4 / 1M tokens

Output

$2 / 1M tokens

Capabilities: tools

Accepts: text

Returns: text

Providers	Context	Input Price	Output Price	Input Formats	Output Formats
mistralai	128K tokens	$0.4 / 1M tokens	$2 / 1M tokens	text	text
openrouter	131072 tokens	$0.4 / 1M tokens	$2 / 1M tokens	text	text

Released Jul 10, 2025Knowledge Cutoff: Jan 11, 2025

Access devstral-medium through LangDB AI Gateway

Recommended

Integrate with mistralai's devstral-medium and 250+ other models through a unified API. Monitor usage, control costs, and enhance security.

Unified API

Cost Optimization

Enterprise Security

Get Started Now

Free tier available • No credit card required

Instant Setup

99.9% Uptime

10,000+Monthly Requests

Available from 2 providers

Provider:

Benchmark Results for devstral-medium

Category Performance Scores:

Academia: Score 38.55 (Top 57% - Rank #197)
Finance: Score 16.30 (Top 87% - Rank #301)
Marketing: Score 40.53 (Top 55% - Rank #190)
Maths: Score 4.70 (Top 92% - Rank #318)
Programming: Score 31.50 (Top 42% - Rank #145)
Science: Score 52.56 (Top 62% - Rank #214)
Writing: Score 40.11 (Top 55% - Rank #190)

Overall Performance: 32.03507936507936 average score across all categories

Detailed Benchmark Scores:

Benchmark	Score	Percentile	Domain
AIME	6.70	Top 69%	Mathematics
AA Coding Index	31.50	Top 42%	Programming
AAII	27.90	Top 49%	General
AA Math Index	4.70	Top 92%	Mathematics
GPQA	49.20	Top 59%	STEM (Physics, Chemistry, Biology)
HLE	3.80	Top 82%	General Knowledge
LiveCodeBench	33.70	Top 43%	Programming
MATH-500	70.70	Top 62%	Mathematics
MMLU-Pro	70.80	Top 56%	General Knowledge
SciCode	29.40	Top 45%	Scientific

GPQA Score: 49.20 - Graduate-level reasoning benchmark

Model Comparison:

Provider: mistralai

Model Type: completions

Context Size: 128000 tokens

Comparing against 345 models in the database

Category Scores

Benchmark Tests

View Other Benchmarks

AIME

6.7

Mathematics

AA Coding Index

31.5

Programming

AAII

27.9

General

AA Math Index

4.7

Mathematics

GPQA

49.2

STEM (Physics, Chemistry, Biology)

HLE

3.8

General Knowledge

LiveCodeBench

33.7

Programming

MATH-500

70.7

Mathematics

MMLU-Pro

70.8

General Knowledge

SciCode

29.4

Scientific

Metric	AIME	AA Coding Index	AAII	AA Math Index	GPQA	HLE	LiveCodeBench	MATH-500	MMLU-Pro	SciCode
Score	6.7	31.5	27.9	4.7	49.2	3.8	33.7	70.7	70.8	29.4

Compare with Similar Models

gemini-2.5-pro-preview

grok-4

Code Examples

Integration samples and API usage

Code Samples for devstral-medium

Python SDK Example:

from openai import OpenAI

client = OpenAI(
    base_url="https://api.us-east-1.langdb.ai/projects/<your_project_id>",
    api_key="<your_api_key>"
)

response = client.chat.completions.create(
    model="devstral-medium",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

TypeScript SDK Example:

import OpenAI from 'openai';

const client = new OpenAI({
    baseURL: "https://api.us-east-1.langdb.ai/projects/<your_project_id>",
    apiKey: "<your_api_key>"
});

const response = await client.chat.completions.create({
    model: "devstral-medium",
    messages: [
        { role: "user", content: "Hello, how are you?" }
    ]
});

console.log(response.choices[0].message.content);

cURL Example:

curl -X POST "https://api.us-east-1.langdb.ai/projects/<your_project_id>/v1/chat/completions" \
  -H "Authorization: Bearer <your_api_key>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "devstral-medium",
    "messages": [
        {"role": "user", "content": "Hello, how are you?"}
    ]
  }'

Model: devstral-medium

Provider: mistralai

API Endpoint: $https://api.us-east-1.langdb.ai

Create API Key

Related Models

Similar models from mistralai

devstral-medium

devstral-medium

Access devstral-medium through LangDB AI Gateway

Category Scores

Benchmark Tests

Compare with Similar Models

Code Examples

Related Models

aion-1.0

aion-1.0-mini

aion-rp-llama-3.1-8b

coder-large

codestral-2501

codestral-2501

devstral-medium by mistralai - AI Model Details, Pricing, and Performance Metrics

devstral-medium

devstral-medium

Access devstral-medium through LangDB AI Gateway

Category Scores

Benchmark Tests

Compare with Similar Models

Code Examples

Related Models

aion-1.0

aion-1.0-mini

aion-rp-llama-3.1-8b

coder-large

codestral-2501

codestral-2501