Model Cards / Mistral AI

Mistral Large 2 Release

model card950 words·4 min read·Aug 20, 2026·Source
Version History
Chaptered summary is still being generated for this document. Showing a heuristic brief in the meantime.
Summary
950-word document condensed to 101 words. Mistral AI · Aug 20, 2026
TL;DR

This latest generation continues to push the boundaries of cost efficiency, speed, and performance. Mistral Large 2 is exposed on la Plateforme and enriched with new features to facilitate building innovative AI applications.

Top benchmarks
BenchmarkVariantScore
MMLUpretrained, accuracy84.0%

Showing top 1 of 13. See full list below.

Capability claim
  • We are releasing Mistral Large 2 under the [Mistral Research License](https://mistral.ai/licenses/MRL-0.1.md), that allows usage and modification for research and non-commercial usages.
Deployment scope
  • available on Vertex AI, in addition to Azure AI Studio, Amazon Bedrock and IBM [watsonx.ai](http://watsonx.ai).

Every italicized passage is a verbatim substring of the source document (checked deterministically after extraction). Field selection is heuristic — some quotes may lack surrounding context and some claims may be absent if no matching pattern appeared. For citation, open the source: original model card · source SHA e4f3f21d34ae · version dated Aug 20, 2026.

Extracted Evaluations(13 results)

Sort by:0/13 rows fully reproducible (0%)
BenchmarkCategoryStateScoreSetupSource
knowledgescored
84.0%
accuracy
pretrainedmissing: shot countmissing: methodmissing: language
self-reported
knowledgementioned
pretrainedmissing: shot countmissing: methodmissing: language
self-reported
/ multilingual
knowledgementioned
pretrainedmissing: shot countmissing: methodmissing: language
self-reported
/ multilingual
knowledgementioned
pretrainedmissing: shot countmissing: methodmissing: language
self-reported
/ multilingual
knowledgementioned
pretrainedmissing: shot countmissing: methodmissing: language
self-reported
/ multilingual
knowledgementioned
pretrainedmissing: shot countmissing: methodmissing: language
self-reported
mathmentioned
8-shotmissing: methodmissing: languagemissing: training state
self-reported
mathmentioned
0-shotmissing: methodmissing: languagemissing: training state
self-reported
othermentioned
missing: shot countmissing: methodmissing: languagemissing: training state
self-reported
othermentioned
missing: shot countmissing: methodmissing: languagemissing: training state
self-reported
othermentioned
missing: shot countmissing: methodmissing: languagemissing: training state
self-reported
othermentioned
missing: shot countmissing: methodmissing: languagemissing: training state
self-reported
othermentioned
average generation length
missing: shot countmissing: methodmissing: languagemissing: training state
self-reported