Mistral Small 3.1
MistralMistralOpen WeightApache 2.0 · Usage Commercial
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.
Date de sortie
2025-03-17
Paramètres
24.0B
Longueur du contexte
33K
Modalités
image, text
Radar de capacités
24
general
25
coding
19
reasoning
31
science
21
agents
60
multimodal
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Classement codage | 458 | 25.0 | AA |
| Classement général | 491 | 28.0 | AA |
| Classement multimodal | 107 | 42.0 | LS |
| Science | 496 | 27.0 | AA |
Scores de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Chat
MT-Bench
0.83 / 100Aut.
IFEvalGoogle Research (2023)
82.9%Aut.
Factuality
SimpleQA
10.4%Aut.
General
Arena Hard
87.6%Aut.
MMLU
80.6%Aut.
TriviaQA
80.5%Aut.
Wild Bench
52.2%Aut.
Language
MMLU-Pro
66.8%Aut.
Math
MATH
69.3%Aut.
Multimodal
MMMU
59.3%Aut.
Reasoning
HumanEvalOpenAI (2021)
88.4%Aut.
MBPP
0.75 / 100Aut.
GPQANYU + Cohere + Anthropic (2023)
46.0%Aut.
Indices d'évaluation AA
(Artificial Analysis)Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))70.7
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))65.9
Gpqa(NYU + Cohere + Anthropic (2023))45.4
Ifbench(Google Research (2023))29.9
Scicode(UIUC + Argonne National Lab (2024))27.8
Coding Index(Artificial Analysis)26.3
Terminalbench V2 126.2
Tau2(Sierra + U Toronto + Vector Institute (2025))25.1
Lcr(Artificial Analysis)22.3
Livecodebench(UC Berkeley + MIT + Cornell (2024))21.2
Aime(MAA (Mathematical Association of America))9.3
Terminalbench Hard(Stanford × Laude Institute (2026))7.6
Tau Banking7.4
Intelligence Index(Artificial Analysis)7.1
Hle(Center for AI Safety + Scale AI (2025))4.3
Math Index(Artificial Analysis)3.7
Aime 25(MAA (Mathematical Association of America))3.7
Scores par catégorie LLM Stats
(LLM Stats (zeroeval))Creativity90
Writing90
Chat80
Instruction Following80
Roleplay80
Structured Output80
Code80
Language70
Legal70
Math70
Reasoning70
Finance70
General70
Healthcare70
Communication70
Physics50
Biology50
Chemistry50
Tarification
Prix d'entrée$0.11 / 1M tokens
Prix de sortie$0.17 / 1M tokens
Prix mixte (3:1)$0.125 / 1M tokens
Vitesse
Tokens/sec0.0
Délai du premier token0.00s
Temps de réponse0.00s
Classement des Prix par Fournisseur
Classement des Prix par Fournisseur
2 fournisseurs
Moins cher: DeepInfraPlus cher: Mistral
FournisseurEntréeSortie
1DeepInfraMoins cher
$0
$0
2MistralPRINCIPAL
$0.11
$0.17
Comparer les prix entre différents fournisseurs API pour ce modèle.