# Cached-input pricing: https://docs.mistral.ai/inference/pricing (2026-09-29)
name = "Mistral Small (latest)"
description = "Efficient Mistral model for fast chat, extraction, and production assistants"
family = "mistral-small"
release_date = "2026-03-16"
last_updated = "2026-03-16"
attachment = true
reasoning = true
temperature = true
knowledge = "2025-06"
tool_call = true
open_weights = true

[[reasoning_options]]
type = "effort"
values = ["none", "high"]

[cost]
input = 0.15
cache_read = 0.015
output = 0.60

[limit]
context = 256_000
output = 256_000

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