Llama 4 Scout
MetaLlamaOpen WeightLlama 4 Community License Agreement
Descripción
Llama 4 Scout is a natively multimodal model capable of processing both text and images. It features a 17 billion activated parameter (109B total) mixture-of-experts (MoE) architecture with 16 experts, supporting a wide range of multimodal tasks such as conversational interaction, image analysis, and code generation. The model includes a 10 million token context window.
Fecha de lanzamiento
2025-04-05
Parámetros
109.0B
Longitud del contexto
1.3M
Modalidades
image, text
Radar de capacidades
28
general
17
coding
33
reasoning
35
science
28
agents
90
multimodal
Rankings
| Dominio | #Posición | Puntuación | Fuente |
|---|---|---|---|
| Ranking de codificación | 531 | 17.0 | AA |
| Ranking general | 451 | 31.0 | AA |
| Ranking multimodal | 88 | 49.0 | LS |
| Ciencia | 486 | 28.0 | AA |
Puntuaciones de benchmarks (LLM Stats)
(LLM Stats (zeroeval))General
MMLU
79.6%Aut.
Language
MMLU-Pro
74.3%Aut.
TydiQA
31.5%Aut.
Math
MGSM
90.6%Aut.
MathVista
70.7%Aut.
MATH
50.3%Aut.
Multimodal
MMMU
69.4%Aut.
Reasoning
ChartQAMasry et al. (2022)
88.8%Aut.
MBPP
0.68 / 100Aut.
GPQANYU + Cohere + Anthropic (2023)
57.2%Aut.
LiveCodeBench
32.8%Aut.
Vision
DocVQADocVQA (2020)
94.4%Aut.
Índices de evaluación AA
(Artificial Analysis)Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))84.4
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))75.2
Gpqa(NYU + Cohere + Anthropic (2023))58.7
Ifbench(Google Research (2023))39.5
Livecodebench(UC Berkeley + MIT + Cornell (2024))29.9
Aime(MAA (Mathematical Association of America))28.3
Lcr(Artificial Analysis)27.7
Scicode(UIUC + Argonne National Lab (2024))21.3
Tau2(Sierra + U Toronto + Vector Institute (2025))15.5
Aime 25(MAA (Mathematical Association of America))14.0
Math Index(Artificial Analysis)14.0
Coding Index(Artificial Analysis)8.2
Intelligence Index(Artificial Analysis)8.1
Hle(Center for AI Safety + Scale AI (2025))3.8
Terminalbench V2 13.7
Tau Banking3.3
Terminalbench Hard(Stanford × Laude Institute (2026))1.5
Terminalbench V4 00.0
Puntuaciones por categoría LLM Stats
(LLM Stats (zeroeval))Image To Text90
Language80
Legal80
Multimodal80
Finance80
Vision80
Math70
Reasoning70
General70
Healthcare70
Physics60
Biology60
Chemistry60
Code30
Precios
Precio de entrada$0.19 / 1M tokens
Precio de salida$0.68 / 1M tokens
Precio mixto (3:1)$0.313 / 1M tokens
Velocidad
Tokens/seg134.4
Retraso del primer token0.67s
Tiempo hasta la respuesta0.67s
Ranking de Precios por Proveedor
Ranking de Precios por Proveedor
6 proveedores
Más barato: DeepInfraMás caro: Meta
ProveedorEntradaSalida
1DeepInfraMás barato
$0
$0
2Helicone
$0.08
$0.3
3NanoGPT
$0.085
$0.46
4OpenRouter
$0.1
$0.3
5Kilo Gateway
$0.1
$0.3
6MetaPRINCIPAL
$0.19
$0.68
Comparar precios entre diferentes proveedores de API para este modelo.