Ministral 3 14B
MistralMistralOpen WeightApache 2.0 · Usage Commercial
Description
A balanced model in the Ministral 3 family, Ministral 3 14B is a powerful, efficient tiny language model with vision capabilities. This model is the instruct post-trained version in FP8, fine-tuned for instruction tasks, making it ideal for chat and instruction based use cases. The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware. Ministral 3 14B can even be deployed locally, capable of fitting in 24GB of VRAM in BF16, and less than 12GB of RAM/VRAM when quantized.
Date de sortie
2025-12-02
Paramètres
14.0B
Longueur du contexte
262K
Modalités
image, text
Radar de capacités
25
general
23
coding
35
reasoning
35
science
31
agents
10
multimodal
Classements
| Domaine | #Rang | Score | Source |
|---|---|---|---|
| Classement codage | 494 | 22.0 | AA |
| Classement général | 512 | 28.0 | AA |
| Science | 474 | 30.0 | AA |
Scores de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Chat
MM-MT-Bench
0.08 / 100Aut.
General
MMLU
79.4%Aut.
TriviaQA
74.9%Aut.
Multilingual MMLU
74.2%Aut.
Wild Bench
68.5%Aut.
Arena Hard
55.1%Aut.
Language
MMLU-Redux
82.0%Aut.
Math
MATH
90.4%Aut.
AIME 2024
89.8%Aut.
AIME 2025
85.0%Aut.
MATH (CoT)
67.6%Aut.
Reasoning
GPQANYU + Cohere + Anthropic (2023)
71.2%Aut.
AGIEval
64.8%Aut.
LiveCodeBench
64.6%Aut.
Indices d'évaluation AA
(Artificial Analysis)Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))69.3
Gpqa(NYU + Cohere + Anthropic (2023))57.2
Livecodebench(UC Berkeley + MIT + Cornell (2024))35.1
Ifbench(Google Research (2023))32.0
Math Index(Artificial Analysis)30.0
Aime 25(MAA (Mathematical Association of America))30.0
Tau2(Sierra + U Toronto + Vector Institute (2025))27.2
Lcr(Artificial Analysis)26.3
Scicode(UIUC + Argonne National Lab (2024))23.8
Coding Index(Artificial Analysis)14.4
Terminalbench V2 19.7
Tau Banking6.6
Intelligence Index(Artificial Analysis)6.0
Hle(Center for AI Safety + Scale AI (2025))4.6
Terminalbench Hard(Stanford × Laude Institute (2026))4.5
Terminalbench V4 00.0
Scores par catégorie LLM Stats
(LLM Stats (zeroeval))Math90
Reasoning70
General60
Creativity60
Writing60
Communication40
Chat30
Multimodal10
Tarification
Prix d'entrée$0.2 / 1M tokens
Prix de sortie$0.2 / 1M tokens
Prix mixte (3:1)$0.2 / 1M tokens
Prix de lecture cache$0.02 / 1M tokens
Vitesse
Tokens/sec82.7
Délai du premier token0.55s
Temps de réponse0.55s
Classement des Prix par Fournisseur
Classement des Prix par Fournisseur
4 fournisseurs
Moins cher: MistralPlus cher: Kilo Gateway
FournisseurEntréeSortie
1MistralPRINCIPAL
$0.2
$0.2
2NanoGPT
$0.2
$0.2
3OpenRouter
$0.2
$0.2
4Kilo Gateway
$0.2
$0.2
Comparer les prix entre différents fournisseurs API pour ce modèle.