NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Mistral Medium 3
Mistral AI logo

Mistral Medium 3

by Mistral AI · Released May 2025

Open SourceMultimodal
37.3
avg score
Rank #211
Compare
Better than 32% of all models
Context
131K tokens (~66 books)
Input $/1M
$0.40
Output $/1M
$2.00
Type
multimodal
License
Open Source
Benchmarks
7 tested
Data as of
About

Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost...

Tested on 7 benchmarks · BenchGecko score 37.3. Top scores: MATH level 5 (81.6%), Lech Mazur Writing (77.3%), GPQA diamond (46.0%).

Looking for similar performance at lower cost?
GPT-5 Nano scores 37.7 (101% as good) at $0.05/1M input · 88% cheaper
Capabilities
math
38.1
#160 globally
knowledge
41.1
#197 globally
general
37.1
#83 globally
Benchmark Scores
Compare All
Tested on 7 benchmarks · Ranked across 3 categories
Score Distribution (all 312 models)
0255075100
▲ You are here
mathCompare math →
MATH level 5

Competition-level math from AMC, AIME, and olympiad problems. Level 5 is the hardest tier, requiring creative problem-solving.

81.6·
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

32.1·
FrontierMath-2025-02-28-Private

Original research-level math problems created by professional mathematicians. Problems are unpublished and cannot be memorized.

0.6·
knowledgeCompare knowledge →
Lech Mazur Writing

Writing quality evaluation by Lech Mazur. Tests prose quality, coherence, and stylistic ability.

77.3·
GPQA diamond

Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.

46.0·
HLE

Humanitys Last Exam. Expert-level questions spanning all academic disciplines, designed to be the hardest knowledge test for AI.

0.0·
generalCompare general →
Dtbench
37.1·
Quick compare:Google DeepMind logovs Gemini 2.5 Flashz-ai logovs GLM 4 32B Anthropic logovs Claude 2
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Mistral AI Mistral Medium
May 2025
37.3
$0.40/M in131Kctx7 benchmarks
Aug 2025
$0.40/M in131Kctx3 benchmarks
Aug 2025
$0.20/M in(-0.20)131Kctx
Apr 2026
25.0-12.3
$1.50/M in(+1.30)262Kctx(+131K)21 benchmarks
Apr 2026
$0.75/M in(-0.75)262Kctx
See the full Mistral Medium family →
Similar Models
Google DeepMind logo
Gemini 2.5 Flash
Google DeepMind
37.3$0.30/1M
z-ai logo
GLM 4 32B
z-ai
37.5$0.10/1M
Anthropic logo
Claude 2
Anthropic
37.7n/a
Links
Info
Mistral AIPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Mistral AI
Source Code
GitHubHugging Face
BenchGecko API
mistral-medium-3
Specifications
  • Typemultimodal
  • Context131K tokens (~66 books)
  • ReleasedMay 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.003
Available On
Mistral AI logoMistral AI$0.40
Categories
mathknowledgegeneral
Learn More
context-windowtransformeropen weightstokens
Share & Export
Tweet
Related Models
Google DeepMind logoGemini 2.5 Flash
37.3
z-ai logoGLM 4 32B
37.5
Anthropic logoClaude 2
37.7
OpenAI logoGPT-5 Nano
37.7
Mistral Medium 3 is an open-source multimodal AI model by Mistral AI, released in May 2025. It has an average benchmark score of 37.3. Context window: 131K tokens.

Related Models

Gemini 2.5 Flash · Google DeepMindGLM 4 32B · z-aiClaude 2 · AnthropicGPT-5 Nano · OpenAIHermes 2 Pro - Llama-3 8B · nousresearch

Benchmarks

MATH level 5Lech Mazur WritingGPQA diamondDtbenchOTIS Mock AIME 2024-2025

Related Pages

Mistral AI · ProviderMistral AI · EconomyAll ModelsCompare ModelsPricingDevelopers · APIContext Window · Glossary

Key facts · as of 2026-10-05

  • Mistral Medium 3 by Mistral AI. BenchGecko score 37.3, rank 211 of 312 scored models (normalized average of public benchmark scores).
  • List price $0.40 input · $2.00 output per 1M tokens (as of 2026-10-05).
  • Sold by 3 providers (as of 2026-10-05): Mistral $0.40 in / $2.00 out · Mistral $0.40 in / $2.00 out · Mistral $0.44 in / $2.20 out. Every provider

How to cite · data as of 2026-10-05

Mistral Medium 3 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/mistral-medium-3

Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP