Mistral Small 3.1 (25.03)
Mistral Small 3.1 (25.03)Compare inon
Mistral Small 3.1 (25.03) measured on 2 of 104 Chat & Knowledge benchmarks2
BenchmarkByScoreRank
Legacy · saturated or no longer reported2
No benchmark matches
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperGPT OSS 120BhighOpen×22.9-60%$0.092BetterGPT-5.6 Lunalow×17.1+200%$0.01
OpenDeepSeek V4 Pro 0813max×25.4+1,220%$0.67Pass back reasoning_content · Chat Completions+4
CheaperQwen3.7 Flashunstated×22.3-63%$0.03 / $0.13CheaperQwen 3.5 Flashunstated×21.7-24%$0.065 / $0.26ModelScoreCostPer taskWhat changes
GPT OSS 120BhighOpen×22.9-60%$0.092Qwen3.7 Flashunstated×22.3-63%$0.03 / $0.13Qwen 3.5 Flashunstated×21.7-24%$0.065 / $0.26GPT-5 Nano (2025-08-07)high×20.9-8%$0.05 / $0.40GPT OSS 20BmediumOpen×16.8-76%$0.018 / $0.09ModelScoreCostPer taskWhat changes
GPT-5.6 Lunalow×17.1+200%$0.01GPT OSS 20BhighOpen×13.1-76%$0.011
Gemini 3.1 Flash Litehigh×20.6+275%$0.06GPT-6 Lunamax×25.4+33%$0.068Function calling only with none · Chat Completions+3
Claude Haiku 5.5high×25+33%$0.079Remove sampling parameters+5ModelScoreCostPer taskWhat changes
DeepSeek V4 Pro 0813maxOpen×25.4+1,220%$0.67Pass back reasoning_content · Chat Completions+4
Kimi K3maxOpen×25+3,900%$2.00Sampling parameters are fixed+3Kimi K2.6unstatedOpen×24.7+1,042%$0.95 / $4Kimi K2.7 CodeunstatedOpen×24.6+1,042%$0.95 / $4Instant mode is not supported · Self-hosted / vLLM / SGLang
GLM 5.3 FlashmaxOpen×24.1+58%$0.28Defaults to max · Self-hosted+4ModelScoreCostPer taskWhat changes
Mistral Large 4high×25.3—$1.36 / $4.18Infinite whitespace stream+4Magistral Small 1unstatedOpen×7.7—$0.70 / $1.16Mistral 7B Instruct v0.3Open-93%—$0.028 / $0.054Mistral Medium 3.5highOpenn/a—$0.50Mistral Small 4thinkingOpenn/a—$0.015Ministral 3 14B 2512non-thinkingOpenn/a—$0.019Ministral 3 3B 2512Openn/a—$0.008Ministral 3 8B 2512Openn/a—$0.011All Mistral models →FAQ
How much does Mistral Small 3.1 (25.03) cost?›
$0.10 per 1M input tokens and $0.30 per 1M output tokens (cheapest host: Google Vertex AI).
Is Mistral Small 3.1 (25.03) free to use?›
There is no free API; the cheapest paid price is $0.10 per 1M input tokens.
How good is Mistral Small 3.1 (25.03), and what is it best at?›
#111 of 133 in Chat & Knowledge.
Is Mistral Small 3.1 (25.03) better than Meta Llama 3 70B Instruct?›
On OTIS Mock AIME 2024 2025, Mistral Small 3.1 (25.03) scores 3.9 vs 4.3 — about the same.
How can I use Mistral Small 3.1 (25.03)?›
Through the Mistral API.
What is Mistral Small 3.1 (25.03)?›
Mistral Small 3.1 (25.03) is a language model from Mistral.
What are the best alternatives to Mistral Small 3.1 (25.03)?›
GPT OSS 120B (cheaper), GPT-5.6 Luna (better), DeepSeek V4 Pro 0813 (open weights).