Mistral AI

Mistral Small 3.1 24B Instruct 2503

Type
language model
Context
32K tokens
Max output
32K
API string
watsonx/mistralai/mistral-small-3-1-24b-instruct-2503

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
IBM watsonx.ai $0.1 $0.3 31 Jul 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score, programming 1,310.1 1,302–1,318
Arena Score, long queries 1,299.32 1,292–1,307
Arena Score in English 1,298.58 1,293–1,304
Arena Score in Spanish 1,281.12 1,258–1,305
Arena Score, hard prompts 1,278.05 1,272–1,284
Arena Score, overall 1,277.89 1,273–1,282
Arena Score in French 1,272.61 1,242–1,303
Arena Score, multi-turn dialogue 1,271.12 1,263–1,280
Arena Score, instruction following 1,265.22 1,258–1,272
Arena Score in Russian 1,264.15 1,251–1,277
Arena Score, mathematics 1,261.51 1,249–1,275
Arena Score, expert questions 1,259.54 1,244–1,275
Arena Score, creative writing 1,253.32 1,244–1,263
Arena Score, working with images 1,136.87 1,128–1,145
Arena Score, text recognition in images 1,131.36 1,121–1,142
Arena Score, understanding diagrams 1,119.69 1,104–1,135