LLM ReferenceLLM Reference

Mistral Large 2

Multimodal

About

Advanced reasoning and coding model, multilingual with native function calling and JSON output

Capabilities

VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution

Providers(3)

Compare all →
ProviderInput (per 1M)Output (per 1M)Type
OpenRouter$0.5$1.5Serverless
IBM watsonx$2$6Serverless
AWS Bedrock$0.48$2.4Serverless

Benchmark Scores(6)

BenchmarkScoreVersionSource
HellaSwag93.810-shotOpen LLM Leaderboard
HumanEval84.8pass@1Open LLM Leaderboard
Massive Multitask Language Understanding84.05-shotOpen LLM Leaderboard
Chatbot Arena1265.0https://lmarena.ai
BFCL38.4https://gorilla.cs.berkeley.edu/leaderboard.html
MMLU PRO69.7https://huggingface.co/spaces/TIGER-Lab/MMLU-Pro

Rankings

Specifications

FamilyMistral
Released2025-11-25
Parameters123B
Context128K
ArchitectureDecoder Only
Knowledge cutoff2025-07
Specializationgeneral
Trainingfinetuning

Created by

Enterprise AI solutions for trust and transparency.

Paris, France
Founded 2023
Website