All times Eastern (ET)Provider pages checked daily at
ProSign inRSS

Mistral Large 2.0

API ID mistral-large-2407

Mistral Large 2.0 was a Mistral model, released July 2024. It had a 128,000-token context window.

A powerfull model with open weights under a MRL license, competing agains SOTA much larger models.

Mistral's description
Status
Retired
Released
July 2024
Shutdown
Mar 30, 2025

mistral-large-2407 stopped serving on March 30, 2025. Recommended replacement: mistral-large-2512.

How to migrate to Mistral Large 3

Benchmarks

Scores as each source published them
or

Comparing with Mistral Large 1.0, the previous model in this line. A score is compared only with the same benchmark from the same source, at the same setting where both have it. Full comparison: prices, limits, features

Arena

ArenaMistral Large 2.0Mistral Large 1.0ChangeRankDateSource
Text
By category
  • Math 1288 rank 263 vs 1244 (+44)
  • Creative writing 1287 rank 253 vs 1211 (+76)
  • Instruction following 1299 rank 264 vs 1238 (+61)
  • Hard prompts 1320 rank 269 vs 1263 (+57)
  • Long prompts 1305 rank 281 vs 1248 (+57)
  • Multi-turn 1297 rank 271 vs 1227 (+70)
  • Expert 1302 rank 260 vs 1230 (+72)
1314 ±41243 ±5+71268 of 415
45,459 votes		Oct 13, 2026source
Text: Coding1354 ±81296 ±9+58266 of 410
7,589 votes		Oct 13, 2026source

Rated as mistral-large-2407; Mistral Large 1.0 as mistral-large-2402. Ratings come from people voting between two anonymous answers; Text is style-controlled, as on arena.ai. Data: Arena (arena.ai), CC BY 4.0.

Epoch AI

BenchmarkMistral Large 2.0Mistral Large 1.0ChangeSettingRunSource
GPQA Diamond49.0%38.8%+10.2 ptsJan 27, 2025source
MATH Level 544.8%24.5%+20.3 ptsJan 27, 2025source
OTIS Mock AIME 2024-20258.5%1.9%+6.6 ptsFeb 25, 2025source

Run independently by Epoch AI. Data: Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

Features

Compare models
FeatureValueSinceSource
Context windowThe most the model can work with in one request, in tokens: your prompt, conversation history and files. A token is about three-quarters of an English word.128,000 tokensin place when tracking began (October 6, 2026)source

Not published by Mistral: Structured outputs, Function calling, Prompt caching, Batch API, Image input, PDF input, Reasoning / thinking, Fine-tuning, Built-in web search, temperature parameter, top_p parameter, logprobs, Max output tokens.

Price history

USD per 1M tokens, standard tier

No price history yet.

Prices

USD per 1M tokens, every type and band, by tier

No published prices.

Changelog

Changes we detected on Mistral's pages

No changes recorded yet.

Last checked · Sources: Mistral models