MistralMistral

Pixtral Large

Mistral's frontier vision and language model

Pixtral Large is Mistral's multimodal flagship, combining strong language reasoning with advanced image understanding. It processes charts, documents, photos, and complex visual scenes with high accuracy, competing with top closed-source vision models.

Try Pixtral Large

Specifications

Context window128k tokens
Input price$2.00 / 1M tokens
Output price$6.00 / 1M tokens
Credits per query1 cr
Released2024-11
LicenseMistral Research

Features

VisionTool useJSON modeFunction callingCode

Benchmark Scores

What do these mean?

Sourced from official model cards and academic papers. Higher is better.

PhD-level science & reasoning

Mathematical olympiad problems

Chatbot Arena human preference ranking

Python coding — pass@1 rate

87%

Competition-level mathematics

General knowledge across 57 academic domains

84%

Multi-turn instruction following (out of 10)

Real-world GitHub issue resolution rate