NEWHow AI models draw the world map · Gecko Tests
Beta
Home/Models/Mistral Large 4
Mistral AI logo

Mistral Large 4

by Mistral AI · Released Oct 2026

Multimodal1M Context
37.2
avg score
Rank #218
Compare
Better than 32% of all models
Context
1.0M tokens (~524 books)
Input $/1M
$0.68
Output $/1M
$2.09
Type
multimodal
License
Proprietary
Benchmarks
15 tested
Data as of
About

Mistral Large 4 is a frontier multimodal (text and image input) model from Mistral AI built for reasoning, coding, and agentic workloads. It offers a 512K-token context window with up...

Tested on 15 benchmarks · BenchGecko score 37.2. Top scores: OTIS Mock AIME 2024-2025 (98.3%), GPQA diamond (84.1%), Artificial Analysis · Long Context Reasoning (81.3%).

Looking for similar performance at lower cost?
GPT-5 Nano scores 37.5 (101% as good) at $0.05/1M input · 93% cheaper
Capabilities
coding
39.4
#142 globally
math
98.3
#1 globally
knowledge
52.0
#115 globally
general
11.7
#194 globally
speed
48.9
#36 globally
Benchmark Scores
Compare All
Tested on 15 benchmarks · Ranked across 5 categories
Score Distribution (all 319 models)
0255075100
▲ You are here
codingCompare coding →
Frontiercode
39.4·
mathCompare math →
OTIS Mock AIME 2024-2025

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

98.3·
knowledgeCompare knowledge →
GPQA diamond

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

84.1·
SimpleQA Verified

Simple factual questions with verified correct answers. Tests accuracy of basic knowledge retrieval. Low scores indicate hallucination.

20.0·
Quick compare:Mistral AI logovs Mistral Medium 3Google DeepMind logovs Gemini 2.5 FlashOpenAI logovs GPT-5 Nano
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Mistral AI Mistral Large
Feb 2024
31.3
$2.00/M in128Kctx14 benchmarks
Nov 2024
35.1+3.8
$2.00/M in131Kctx(+3K)7 benchmarks
Nov 2024
35.8+0.7
$2.00/M in131Kctx11 benchmarks
Dec 2025
$0.50/M in(-1.50)262Kctx(+131K)
Dec 2025
$0.25/M in(-0.25)262Kctx
Oct 2026
37.2+1.4
$0.68/M in(+0.43)1.0Mctx(+786K)
See the full Mistral Large family →
Recently Happened
New model: Mistral Large 4 by Mistral AI
Oct 8, 2026
Similar Models
Mistral AI logo
Mistral Medium 3
Mistral AI
37.2$0.40/1M
Google DeepMind logo
Gemini 2.5 Flash
Google DeepMind
37.3$0.30/1M
OpenAI logo
GPT-5 Nano
OpenAI
37.5$0.05/1M
Links
Info
Mistral AIPricing explorerDevelopers · API
Research
Technical Report
Documentation
API DocsPlayground
Community
@Mistral AI
BenchGecko API
mistral-large-4-0
Specifications
  • Typemultimodal
  • Context1.0M tokens (~524 books)
  • ReleasedOct 2026
  • LicenseProprietary
  • StatusActive
  • Cost / Message~$0.003
Available On
Mistral AI logoMistral AI$0.68
Categories
codingmathknowledgegeneralspeed
Learn More
context-windowtransformertokens
Share & Export
Tweet
Related Models
Mistral AI logoMistral Medium 3
37.2
Google DeepMind logoGemini 2.5 Flash
37.3
OpenAI logoGPT-5 Nano
37.5
z-ai logoGLM 4 32B
37.5
Mistral Large 4 is a proprietary multimodal AI model by Mistral AI, released in October 2026. It has an average benchmark score of 37.2. Context window: 1M tokens.

Related Models

Mistral Medium 3 · Mistral AIGemini 2.5 Flash · Google DeepMindGPT-5 Nano · OpenAIGLM 4 32B · z-aiClaude 2 · Anthropic

Benchmarks

OTIS Mock AIME 2024-2025GPQA diamondArtificial Analysis · Long Context ReasoningArtificial Analysis · MMMU ProArtificial Analysis · SciCode

Related Pages

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

Key facts · as of 2026-10-15

  • Mistral Large 4 by Mistral AI. BenchGecko score 37.2, rank 218 of 319 scored models (normalized average of public benchmark scores).
  • List price $0.68 input · $2.09 output per 1M tokens (as of 2026-10-15).
  • Sold by 3 providers (as of 2026-10-15): Mistral $0.68 in / $2.09 out · Mistral $0.68 in / $2.09 out · Mistral $0.75 in / $2.30 out. Every provider
  • Gecko Tests: Who Are You B (Knows who made it).

How to cite · data as of 2026-10-15

Mistral Large 4 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-15. https://benchgecko.ai/model/mistral-large-4-0

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