Granite 4.2 3B
IBMOpen WeightApache 2.0 · Uso Comercial
Descripción
IBM Granite 4.2 3B is the compact dense decoder-only reasoning LLM in the Granite 4.2 family. It supports native think, non-think, and low-effort modes, tool calling, and a 128K native context window with extension to 512K. Unlike the 8B and 30B, it did not receive the extra agentic RL block. Open weights on Hugging Face under Apache 2.0. Text-only (not multimodal).
Fecha de lanzamiento
2026-08-25
Parámetros
3.0B
Longitud del contexto
—
Modalidades
text
Radar de capacidades
9
general
19
coding
56
reasoning
35
science
50
agents
0
multimodal
Rankings
| Dominio | #Posición | Puntuación | Fuente |
|---|---|---|---|
| Capacidad agéntica | 112 | 35.0 | LS |
| Ranking de codificación | 389 | 23.0 | AA |
| Ranking general | 543 | 11.0 | AA |
| Ciencia | 364 | 31.0 | AA |
Puntuaciones de benchmarks (LLM Stats)
(LLM Stats (zeroeval))Agents
TAU3-Bench
45.8%Aut.
General
Arena-Hard v2
35.0%Aut.
Instruction Following
IFBench
74.3%Aut.
Language
MMLU-Pro
67.8%Aut.
Long Context
RULER 64k
67.5%Aut.
RULER 128k
55.3%Aut.
Math
AIME 2025
78.3%Aut.
HMMT25
66.7%Aut.
Reasoning
LiveCodeBench v6
69.7%Aut.
GPQANYU + Cohere + Anthropic (2023)
54.8%Aut.
ProfBench
32.1%Aut.
SciCode
24.1%Aut.
Tool Calling
BFCL-V4
52.4%Aut.
Índices de evaluación AA
(Artificial Analysis)Gpqa(NYU + Cohere + Anthropic (2023))55.9
Scicode(UIUC + Argonne National Lab (2024))25.3
Lcr(Artificial Analysis)24.3
Coding Index(Artificial Analysis)17.5
Terminalbench V2 113.9
Intelligence Index(Artificial Analysis)9.1
Hle(Center for AI Safety + Scale AI (2025))6.6
Tau Banking5.6
Terminalbench V4 00.0
Puntuaciones por categoría LLM Stats
(LLM Stats (zeroeval))Instruction Following70
Language70
Legal70
Finance70
Healthcare70
Long Context60
Math60
General60
Reasoning50
Tool Calling50
Physics40
Agents40
Biology40
Chemistry40
Knowledge30
Creativity30
Writing30
Code20
Precios
Precio de entrada$0.03 / 1M tokens
Precio de salida$0.12 / 1M tokens
Precio mixto (3:1)$0.052 / 1M tokens
Velocidad
Tokens/seg39.4
Retraso del primer token0.36s
Tiempo hasta la respuesta51.17s
Ranking de Precios por Proveedor
Ranking de Precios por Proveedor
2 proveedores
Más barato: DeepInfraMás caro: IBM
ProveedorEntradaSalida
1DeepInfraMás barato
$0
$0
2IBMPRINCIPAL
$0.03
$0.12
Comparar precios entre diferentes proveedores de API para este modelo.