modelbenchmark.io

Mistral Small 3.1 24B (2503)

Mistral · mistral-small-3-1-24b-instruct

Compare

Building upon Mistral Small 3 (2501), Mistral Small 3.1 (2503) adds state-of-the-art vision understanding and enhances long context capabilities up to 128k tokens without compromising text performance. With 24 billion parameters, this model achieves top-tier capabilities in both text and vision tasks.

Specification

most-agreed values

Context
131K
Max output
16K
Released
2025-04-15
Knowledge cutoff
2024-06
Retires
Open weights
yes
Input
text, image
Output
text

Price

US dollars per million tokens · most-agreed

Input
$0.106
Output
$0.318
Cache read
$0.05

Rate limits

Requests per minute
300

Available from 6 hosts

HostIn $/MOut $/MCache rdCache wrContextOutputRetires
nano-gpt$0.10$0.30$0.05128K102K
watsonx$0.106$0.318131K16K
cloudflare$0.351$0.555128K128K
cloudflare-workers-ai · cf$0.351$0.555128K128K
kilo$0.351$0.555128K102K
openrouter$0.351$0.555
128K
131K
102K
131K

Price history

append-only observations · a listing writes a row only when its price moves

DateHostField$/M
2026-09-02openrouterInput$0.351
2026-09-06openrouterInput$0.351
2026-09-02watsonxInput$0.106
2026-09-06watsonxInput$0.106
2026-09-02openrouterOutput$0.555
2026-09-06openrouterOutput$0.555
2026-09-02watsonxOutput$0.318
2026-09-06watsonxOutput$0.318

More from Mistral

most-hosted first