name = "Pixtral Large (25.02)"
description = "Mistral vision-language model for image understanding and multimodal chat"
family = "pixtral"
release_date = "2025-04-08"
last_updated = "2025-04-08"
attachment = true
reasoning = false
temperature = true
tool_call = true
open_weights = false

[limit]
context = 128_000
output = 8_192

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