โ† Back to Model Hub/SIDE-BY-SIDE REVIEW
SHARE THIS:

Mistral Large 2 vs Mistral Large 3

How do these models stack up? Below is an expert side-by-side comparison of specifications, context window capacity, live pricing per million tokens, and standardized benchmark scores for Mistral Large 2 and Mistral Large 3.

Mistral

Mistral Large 2

Mistral Large 2 is Mistral AI's second-generation flagship commercial model, delivering substantial improvements over the original Mistral Large in reasoning, tool use, multilingual capabilities, and instruction following. With a 1-million-token context window, it processes entire codebases and lengthy documents in a single pass. Mistral Large 2 excels at complex agentic workflows, structured JSON output, and native function calling โ€” making it one of the top choices for European enterprises requiring GDPR-compliant AI infrastructure.

View Full Specs
Mistral

Mistral Large 3

Mistral's flagship commercial model, boasting multilingual support and advanced coding and math skills. Designed for complex reasoning and enterprise tasks that require high compliance.

View Full Specs

Technical Specifications

SpecificationMistral Large 2Mistral Large 3
ProviderMistralMistral
Context Window1,048,576 tokens262,144 tokens
Agent Suitability93/10091/100
Time to First Token (TTFT)270 ms250 ms
Deployment Modelmanaged apimanaged api
Production Stabilitystablestable
API AvailableYesYes
Released Date2025-07-242024-07-24

API Pricing Comparison

Input Price per Million Tokens

Mistral Large 2

$0.60

Mistral Large 3

$0.50

Output Price per Million Tokens

Mistral Large 2

$1.80

Mistral Large 3

$1.50

Want to test both models live?

Run side-by-side prompt prompts in our dynamic Sandbox. Check execution speeds, latency metrics, and compute actual costs in real-time.

Benchmark Performance Metrics

Scores show the raw performance percentages verified across key evaluation suites. Higher bars indicate superior accuracy and capability in that domain.

MMLUGeneral knowledge & multi-task understanding
N/Avs8680.0%
Mistral Large 2
Mistral Large 3
HumanEvalPython coding & logic synthesis
N/Avs8820.0%
Mistral Large 2
Mistral Large 3
MATHComplex mathematical problem solving
N/Avs7950.0%
Mistral Large 2
Mistral Large 3
GPQAGraduate-level expert reasoning
N/Avs5850.0%
Mistral Large 2
Mistral Large 3
HellaSwagCommonsense reasoning and inference
N/Avs9000.0%
Mistral Large 2
Mistral Large 3
MT-BenchMulti-turn conversation flow quality
N/Avs900.0%
Mistral Large 2
Mistral Large 3

Mistral Large 2 Quirks & Gotchas

  • โ–ธExcellent European data sovereignty โ€” GDPR-compliant infrastructure
  • โ–ธ1M context window enables full codebase analysis in a single pass

Mistral Large 3 Quirks & Gotchas

  • โ–ธTop-tier structured JSON output โ€” best multilingual function calling
  • โ–ธLe Chat offers free consumer tier with same underlying model