Mistral Large 3
MistralMistralOpen WeightApache 2.0 · Uso Comercial
Descripción
Mistral Large 3 is a state-of-the-art general-purpose Multimodal granular Mixture-of-Experts model with 41B active parameters and 675B total parameters trained from scratch with 3000 H200s. This model is the instruct post-trained version, fine-tuned for instruction tasks, making it ideal for chat, agentic and instruction based use cases. Designed for reliability and long-context comprehension - It is engineered for production-grade assistants, retrieval-augmented systems, scientific workloads, and complex enterprise workflows.
Fecha de lanzamiento
2025-12-02
Parámetros
675.0B
Longitud del contexto
262K
Modalidades
image, text
Radar de capacidades
30
general
31
coding
43
reasoning
44
science
39
agents
85
multimodal
Rankings
| Dominio | #Posición | Puntuación | Fuente |
|---|---|---|---|
| Ranking de codificación | 400 | 33.0 | AA |
| Ranking general | 396 | 34.0 | AA |
| Ciencia | 370 | 40.0 | AA |
Puntuaciones de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Chat
MM-MT-Bench
84.90 / 100Aut.
Factuality
SimpleQA
23.8%Aut.
General
TriviaQA
74.9%Aut.
Wild Bench
68.5%Aut.
Arena Hard
55.1%Aut.
Language
MMMLU
85.5%Aut.
MMLU-Redux
82.0%Aut.
Math
MATH
90.4%Aut.
MATH (CoT)
67.6%Aut.
AMC_2022_23
52.0%Aut.
Reasoning
GPQANYU + Cohere + Anthropic (2023)
43.9%Aut.
LiveCodeBench
34.4%Aut.
Índices de evaluación AA
(Artificial Analysis)Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))80.7
Gpqa(NYU + Cohere + Anthropic (2023))68.0
Livecodebench(UC Berkeley + MIT + Cornell (2024))46.5
Math Index(Artificial Analysis)38.0
Aime 25(MAA (Mathematical Association of America))38.0
Scicode(UIUC + Argonne National Lab (2024))36.6
Ifbench(Google Research (2023))36.2
Lcr(Artificial Analysis)36.0
Tau2(Sierra + U Toronto + Vector Institute (2025))24.6
Coding Index(Artificial Analysis)20.1
Terminalbench Hard(Stanford × Laude Institute (2026))15.9
Terminalbench V2 112.0
Intelligence Index(Artificial Analysis)9.3
Tau Banking5.8
Hle(Center for AI Safety + Scale AI (2025))4.2
Puntuaciones por categoría LLM Stats
(LLM Stats (zeroeval))Multimodal85
Communication43
Chat43
General11
Language80
Math80
Reasoning70
Creativity60
Writing60
Precios
Precio de entrada$0.5 / 1M tokens
Precio de salida$1.5 / 1M tokens
Precio mixto (3:1)$0.75 / 1M tokens
Precio de lectura caché$0.05 / 1M tokens
Velocidad
Tokens/seg77.6
Retraso del primer token0.78s
Tiempo hasta la respuesta0.78s
Ranking de Precios por Proveedor
Ranking de Precios por Proveedor
8 proveedores
Más barato: MistralMás caro: 302.AI
ProveedorEntradaSalida
1MistralPRINCIPAL
$0.5
$1.5
2OpenRouter
$0.5
$1.5
3Kilo Gateway
$0.5
$1.5
4DevPass (LLM Gateway)
$0.5
$1.5
5Pioneer
$0.5
$1.5
6Opper
$0.5
$1.5
7Cortecs
$0.613
$1.838
8302.AI
$1.1
$3.3
Comparar precios entre diferentes proveedores de API para este modelo.