Ministral 3 3B
MistralMistralOpen WeightApache 2.0 · Uso Comercial
Descripción
The smallest model in the Ministral 3 family, Ministral 3 3B 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 3B can even be deployed locally, fitting in 16GB of VRAM in BF16, and less than 8GB of RAM/VRAM when quantized.
Fecha de lanzamiento
2025-12-02
Parámetros
3.0B
Longitud del contexto
131K
Modalidades
image, text
Radar de capacidades
19
general
13
coding
24
reasoning
23
science
21
agents
10
multimodal
Rankings
| Dominio | #Posición | Puntuación | Fuente |
|---|---|---|---|
| Ranking de codificación | 571 | 12.0 | AA |
| Ranking general | 566 | 22.0 | AA |
| Ciencia | 601 | 17.0 | AA |
Puntuaciones de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Chat
MM-MT-Bench
0.08 / 100Aut.
General
MMLU
70.7%Aut.
Multilingual MMLU
65.2%Aut.
TriviaQA
59.2%Aut.
Wild Bench
56.8%Aut.
Arena Hard
30.5%Aut.
Language
MMLU-Redux
73.5%Aut.
Math
MATH
83.0%Aut.
AIME 2024
77.5%Aut.
AIME 2025
72.1%Aut.
MATH (CoT)
60.1%Aut.
Reasoning
LiveCodeBench
54.8%Aut.
GPQANYU + Cohere + Anthropic (2023)
53.4%Aut.
AGIEval
51.1%Aut.
Índices de evaluación AA
(Artificial Analysis)Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))52.4
Gpqa(NYU + Cohere + Anthropic (2023))35.8
Ifbench(Google Research (2023))26.8
Tau2(Sierra + U Toronto + Vector Institute (2025))24.9
Livecodebench(UC Berkeley + MIT + Cornell (2024))24.7
Math Index(Artificial Analysis)22.0
Aime 25(MAA (Mathematical Association of America))22.0
Lcr(Artificial Analysis)17.0
Scicode(UIUC + Argonne National Lab (2024))15.3
Hle(Center for AI Safety + Scale AI (2025))5.4
Intelligence Index(Artificial Analysis)4.8
Coding Index(Artificial Analysis)4.8
Tau Banking4.7
Terminalbench Hard(Stanford × Laude Institute (2026))0.0
Terminalbench V2 10.0
Terminalbench V4 00.0
Puntuaciones por categoría LLM Stats
(LLM Stats (zeroeval))Math80
Reasoning60
General40
Communication30
Creativity30
Writing30
Chat20
Multimodal10
Precios
Precio de entrada$0.1 / 1M tokens
Precio de salida$0.1 / 1M tokens
Precio mixto (3:1)$0.1 / 1M tokens
Precio de lectura caché$0.01 / 1M tokens
Velocidad
Tokens/seg236.4
Retraso del primer token0.56s
Tiempo hasta la respuesta0.56s
Ranking de Precios por Proveedor
Ranking de Precios por Proveedor
4 proveedores
Más barato: MistralMás caro: Kilo Gateway
ProveedorEntradaSalida
1MistralPRINCIPAL
$0.1
$0.1
2NanoGPT
$0.1
$0.1
3OpenRouter
$0.1
$0.1
4Kilo Gateway
$0.1
$0.1
Comparar precios entre diferentes proveedores de API para este modelo.