Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. Mistral

Mistral Large 4

Mistral Large 4 is an open-weight multimodal mixture-of-experts model with 1.05T total and 49B active parameters, built for reasoning, coding, tool use, and structured output. It supports a 512K-token context window.

  • Reasoning
  • Vision
  • Tool Calling
  • Structured Output

Added Oct 6, 2026

Pricing

Auto routing · per 1M tokens
Input
$0.68
Output
$2.09
Cache read
$0.07
Compare provider prices

Specifications

Context window
524.3K
Max output
N/A
Parameters
1.1T

Benchmarks

Sourced from LMArena.

Arena Score

1420.4

Overall Rank

#134 / 421

Votes

6,423

Confidence Interval

1412.0 - 1428.9

Category Scores

Coding

#110 / 416

1,500 votes

1483.5

Math

#68 / 404

305 votes

1458.9

Longer Query

#143 / 399

2,910 votes

1420.2

Creative Writing

#173 / 419

1,569 votes

1358.8

Instruction Following

#118 / 421

2,318 votes

1418.0

Hard Prompts

#115 / 421

4,151 votes

1452.5

Additional Categories
18

Spanish

#76 / 304

241 votes

1450.0

Industry Mathematical

#82 / 402

335 votes

1455.2

Industry Business And Management And Financial Operations

#95 / 414

992 votes

1443.5

Industry Software And It Services

#103 / 421

2,312 votes

1475.8

Expert

#106 / 371

625 votes

1464.2

Chinese

#108 / 395

527 votes

1471.9

Hard Prompts English

#111 / 419

1,474 votes

1460.0

Industry Legal And Government

#113 / 393

424 votes

1440.1

Industry Life And Physical And Social Science

#117 / 419

984 votes

1448.9

Russian

#117 / 385

1,033 votes

1422.2

Multi Turn

#122 / 419

953 votes

1429.3

French

#125 / 301

214 votes

1438.5

English

#127 / 421

2,329 votes

1434.9

Exclude Ties

#134 / 421

4,818 votes

1410.8

Non English

#134 / 421

4,094 votes

1404.6

Industry Entertainment And Sports And Media

#154 / 419

1,735 votes

1374.6

Industry Writing And Literature And Language

#162 / 420

2,054 votes

1382.0

Industry Medicine And Healthcare

#172 / 391

411 votes

1414.1

Published 2026-11-11 · Matched as Mistral Large 4 · CC BY 4.0

LMArena Dataset

Leaderboard results

BenchmarkScoreRank on NanoGPT
Epoch AI Benchmarking Hub

Epoch AI data, CC BY 4.0 · Updated Nov 16, 2026

Epoch Capabilities Index

154.6

#46 of 115

FrontierMath v2 (Tiers 1–3)

58.6%

#45 of 83

GPQA Diamond

88.1%

#54 of 119

SimpleQA Verified

20.0%

#73 of 82

FrontierMath v2 (Tier 4)

26.8%

#43 of 67

OTIS Mock AIME 2024–2025

98.3%

#29 of 121

MirrorCode

40.0%

#25 of 39

Chess Puzzles

33.0%

#49 of 82

Mystery Game Puzzles

17.0%

#58 of 77

Filtered and ranked to models on NanoGPT.

Providers

Auto routing is available for this model. Explicit provider selection is not available.

Auto
Auto
Recommended
$0.68$2.09$0.07N/AN/A
Batch API
ProviderInputOutputCached input
Auto
Automistralai/mistral-large-4
$0.68/M$2.09/M$0.07/M

Batch rows for this model use normal routing and cost the same as standard API requests. Set provider in a row to choose a provider.

Reliability

Related text models

Compare Mistral Large 4 with similar models from the same provider or model family.

Mistral Medium 3.5 Thinking

mistralai/mistral-medium-3.5:thinking

Mistral Medium 3.5 with reasoning enabled by default (reasoning_effort=high), for complex coding, agentic, and multi-step reasoning prompts.

Mistral Medium 3.5

mistralai/mistral-medium-3.5

Mistral Medium 3.5 is a 128B dense open-weights flagship model for instruction-following, reasoning, coding, long-horizon agentic work, tool use, structured output, and multimodal prompts. It supports a 256k context window and configurable reasoning effort.

Mistral Large 2411

mistralai/mistral-large

Upgrade to Mistral's flagship model. It is fluent in English, French, Spanish, German, and Italian, with high grammatical accuracy, with a long context window.

Mistral Small 3 24B (2501)

mistralai/mistral-small-24b-instruct-2501

Mistral Small 3 24B (2501) hosted by IONOS in Berlin, Germany. Zero data retention.

Mistral Code Agent Latest

mistral-code-agent-latest

Mistral Code Agent Latest is Mistral's direct API alias for devstral-2512, an agentic coding model built for autonomous software engineering, tool use, and long-running code tasks.

Mistral Code Latest

mistral-code-latest

Mistral Code Latest is Mistral's direct API alias for codestral-2508, a low-latency coding model for code generation, completion, fill-in-the-middle workflows, function calling, and structured output.