Mistral Large 3

Mistral Large 3

Mistral · Released Dec 2, 2025 Professional
Intelligence #14 / 590
82.0 Our Score
Speed #197 / 279
55.9 tokens / sec
Input #383 / 592
$0.500 per 1M tokens
Output #363 / 592
$1.50 per 1M tokens
Context
— Not reported

Analysis Summary

Mistral Large 3 is Mistral's flagship large model in this family, with an intelligence index of 15.9 and a coding index of 20.1. Its MMLU Pro of 80.7% and GPQA of 68.0% reflect strong general knowledge and reasoning for a model at this price tier. Livecodebench at 46.5% and scicode at 36.2% indicate functional coding capability suitable for many business development tasks.

For businesses, Mistral Large 3 is well-suited to content generation, SEO copy, structured document drafting, and moderate coding assistance. Its instruction-following benchmark (IFBench 36.2%) and LCR of 34.7% are adequate for most content workflows, though it falls short of frontier models on complex reasoning or agentic tasks. The agentic index of 20.2 limits its use in autonomous pipelines.

At $0.50 input and $1.50 output per million tokens, it offers strong price-performance for content-heavy workloads. Teams that need a capable, affordable model for writing, summarisation, and light coding will find it a practical fit, particularly where frontier-level reasoning is not required.

Assessed June 30, 2026

Editorial notes

Mistral Large 3 delivers strong general knowledge and solid coding at a very competitive $0.50 input price, making it a practical choice for content and SEO workflows on a budget.

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence3.1Technical3Value6.5Content4
Intelligence 3.1/10
Technical 3/10
Content 4/10
Value 6.5/10

How Mistral Large 3 compares

Mistral Large 3 ranks #167 of 385 AI models we track for overall intelligence, #90 of 139 for coding, #179 of 293 for agentic tasks. At $0.50 per million input tokens it is cheaper than 35% of comparable models.

Performance Indices

Source: Artificial Analysis

15.9 Intelligence Index
20.1 Coding Index
20.2 Agentic Index
38 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 68% Graduate-level scientific reasoning
HLE 4.1% Humanity's Last Exam
MMLU Pro 80.7% Multi-task language understanding
AIME 2025 38% Competition mathematics (2025)
SciCode 36.2% Scientific computing

Technical

LiveCodeBench 46.5% Live coding evaluation
TerminalBench Hard 15.9% Agentic terminal tasks
τ²-Bench 24.6% Conversational agent benchmark

Content

IFBench 36.2% Instruction following
LCR 34.7% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does Mistral Large 3 stack up?

Compare side-by-side with other professional models.

Compare Models

Model Information

ProviderMistral
Release Date December 2, 2025
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.50 $0.000500
Output $1.50 $0.001500

Leaderboard Categories

Frequently asked questions about Mistral Large 3

How much does Mistral Large 3 cost?

Mistral Large 3 costs $0.50 per million input tokens and $1.50 per million output tokens.

Is Mistral Large 3 good for coding?

On our coding benchmark index, Mistral Large 3 ranks #90 of 139 models, placing it in the broader range of the field for code generation and debugging.

Who created Mistral Large 3?

Mistral Large 3 is developed by Mistral and was released on December 2, 2025.