# Cached-input pricing: https://docs.mistral.ai/inference/pricing (2026-09-29)
name = "Mistral Large (latest)"
description = "Flagship Mistral model for advanced reasoning, coding, and multilingual work"
family = "mistral-large"
release_date = "2024-11-01"
last_updated = "2025-12-02"
attachment = true
reasoning = false
temperature = true
knowledge = "2024-11"
tool_call = true
open_weights = true

[cost]
input = 0.5
cache_read = 0.05
output = 1.5

[limit]
context = 262_144
output = 262_144

[modalities]
input = ["text", "image"]
output = ["text"]