Skip to main content

Mistral Small 3.1

MistralMistralOpen WeightApache 2.0 · Commercial OK

Description

Mistral Small 3 is a 24B-parameter LLM licensed under Apache-2.0. It focuses on low-latency, high-efficiency instruction following, maintaining performance comparable to larger models. It provides quick, accurate responses for conversational agents, function calling, and domain-specific fine-tuning. Suitable for local inference when quantized, it rivals models 2–3× its size while using significantly fewer compute resources.

Release Date
2025-03-17
Parameters
24.0B
Context Length
33K
Modalities
image, text

Capability Radar

70
general
80
coding
70
reasoning
43
scienceest.
73
agents
60
multimodal

Science is estimated from LLM Stats science scores or reasoning when no dedicated science benchmarks are available.

Rankings

No ranking data available

Benchmark Scores (LLM Stats)

(LLM Stats (zeroeval))

Chat

MT-Bench0.83 / 100SR
IFEvalGoogle Research (2023)82.9%SR

Factuality

SimpleQA10.4%SR

General

Arena Hard87.6%SR
MMLU80.6%SR
TriviaQA80.5%SR
Wild Bench52.2%SR

Language

MMLU-Pro66.8%SR

Math

MATH69.3%SR

Multimodal

MMMU59.3%SR

Reasoning

HumanEvalOpenAI (2021)88.4%SR
MBPP0.75 / 100SR
GPQANYU + Cohere + Anthropic (2023)46.0%SR

AA Evaluation Indices

(Artificial Analysis)

No AA evaluation data available

LLM Stats Category Scores

(LLM Stats (zeroeval))
Creativity
90
Writing
90
Chat
80
Instruction Following
80
Roleplay
80
Structured Output
80
Code
80
Language
70
Legal
70
Math
70
Reasoning
70
Finance
70
General
70
Healthcare
70
Communication
70
Physics
50
Biology
50
Chemistry
50

Pricing

Input Price$0.08 / 1M tokens
Output Price$0.135 / 1M tokens
Blended Price (3:1)$0.0575 / 1M tokens

Speed

Tokens/sec60.0
Time to First Token0.92s
Time to Answer0.92s

Provider Price Ranking

Provider Price Ranking

2 providers

Cheapest: DeepInfraMost Expensive: Mistral
ProviderInputOutput
1DeepInfraCheapest
$0.05
$0.08
2MistralPRIMARY
$0.08
$0.135

Compare pricing across different API providers for this model.

External Sources