Granite 4.2 3B

IBM開源權重Apache 2.0 · 商用許可

描述

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).

發布日期

2026-08-25

參數規模

3.0B

上下文長度

—

支援模態

text

能力雷達圖

general

coding

reasoning

science

agents

multimodal

排行榜排名

領域	#排名	分數	來源
智慧體能力模型榜	112	35.0	LS
程式碼能力榜	389	23.0	AA
通用能力榜	543	11.0	AA
科學能力	364	31.0	AA

基準測試分數 (LLM Stats)

(LLM Stats (zeroeval))

Agents

TAU3-Bench

45.8%自報

General

Arena-Hard v2

35.0%自報

Instruction Following

IFBench

74.3%自報

Language

MMLU-Pro

67.8%自報

Long Context

RULER 64k

67.5%自報

RULER 128k

55.3%自報

Math

AIME 2025

78.3%自報

HMMT25

66.7%自報

Reasoning

LiveCodeBench v6

69.7%自報

GPQANYU + Cohere + Anthropic (2023)

54.8%自報

ProfBench

32.1%自報

SciCode

24.1%自報

Tool Calling

BFCL-V4

52.4%自報

AA 評測指數

(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 1

13.9

Intelligence Index(Artificial Analysis)

9.1

Hle(Center for AI Safety + Scale AI (2025))

6.6

Tau Banking

5.6

Terminalbench V4 0

0.0

LLM Stats 分類評分

(LLM Stats (zeroeval))

Instruction Following

Language

Legal

Finance

Healthcare

Long Context

Math

General

Reasoning

Tool Calling

Physics

Agents

Biology

Chemistry

Knowledge

Creativity

Writing

Code

定價

輸入價格$0.03 / 1M tokens

輸出價格$0.12 / 1M tokens

混合價格(3:1)$0.052 / 1M tokens

速度

Tokens/秒39.4

首Token延遲0.36s

首回答延遲51.17s

供應商價格排行

2 個供應商

最便宜: DeepInfra最貴: IBM

供應商輸入輸出

1DeepInfra最便宜

2IBM主要

$0.03

$0.12

比較該模型在不同 API 供應商之間的定價。

外部連結

LLM Stats Artificial Analysis