Granite 4.0 Micro
IBMOpen WeightApache 2.0 · Uso Comercial
Descripción
A preliminary version of the smallest model in the upcoming Granite 4.0 family, released May 2025. It utilizes a novel hybrid Mamba-2/Transformer, fine-grained mixture of experts (MoE) architecture (7B total parameters, 1B active at inference). This preview version is partially trained (2.5T tokens) but demonstrates significant memory efficiency and performance potential, validated for at least 128K context length without positional encoding.
Fecha de lanzamiento
2025-09-22
Parámetros
7.0B
Longitud del contexto
131K
Modalidades
text
Radar de capacidades
17
general
18
coding
11
reasoning
25
science
13
agents
0
multimodal
Rankings
| Dominio | #Posición | Puntuación | Fuente |
|---|---|---|---|
| Ranking de codificación | 586 | 10.0 | AA |
| Ranking general | 606 | 17.0 | AA |
| Ciencia | 589 | 17.0 | AA |
Puntuaciones de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Chat
IFEvalGoogle Research (2023)
63.0%Aut.
General
MMLU
60.4%Aut.
TruthfulQA
58.1%Aut.
AlpacaEval 2.0
35.2%Aut.
Arena Hard
26.7%Aut.
PopQA
22.9%Aut.
Math
GSM8k
70.1%Aut.
Reasoning
HumanEvalOpenAI (2021)
82.4%Aut.
HumanEval+
78.3%Aut.
BIG-Bench Hard
55.7%Aut.
DROP
46.2%Aut.
Safety
AttaQ
86.1%Aut.
Índices de evaluación AA
(Artificial Analysis)Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))44.7
Gpqa(NYU + Cohere + Anthropic (2023))33.6
Ifbench(Google Research (2023))24.8
Livecodebench(UC Berkeley + MIT + Cornell (2024))18.0
Tau2(Sierra + U Toronto + Vector Institute (2025))12.6
Lcr(Artificial Analysis)7.0
Math Index(Artificial Analysis)6.0
Aime 25(MAA (Mathematical Association of America))6.0
Intelligence Index(Artificial Analysis)5.1
Hle(Center for AI Safety + Scale AI (2025))5.0
Terminalbench Hard(Stanford × Laude Institute (2026))1.5
Puntuaciones por categoría LLM Stats
(LLM Stats (zeroeval))Safety90
Code80
Instruction Following60
Language60
Legal60
Math60
Structured Output60
Finance60
General60
Healthcare60
Reasoning50
Chat40
Creativity30
Writing30
Precios
Precio de entradaGratis
Precio de salidaGratis
Precio mixto (3:1)Gratis
Velocidad
Tokens/seg0.0
Retraso del primer token0.00s
Tiempo hasta la respuesta0.00s
Ranking de Precios por Proveedor
Ranking de Precios por Proveedor
2 proveedores
Más barato: OpenRouterMás caro: Kilo Gateway
ProveedorEntradaSalida
1OpenRouterMás barato
$0.017
$0.112
2Kilo Gateway
$0.017
$0.112
Comparar precios entre diferentes proveedores de API para este modelo.