IBM Granite 4.2 8B

Name: IBM Granite 4.2 8B
Author: IBM

IBM

IBM Granite 4.2 8B is a mid-size dense decoder-only reasoning LLM in the Granite 4.2 family with native think, non-think, and low-effort modes plus tool calling. It offers a 128K native context window extensible to 512K and received additional agentic reinforcement learning for multi-step tool use. Open weights on Hugging Face under Apache 2.0; text-only.

Key Specifications

Parameters

8.0B

Context

Release Date

August 25, 2026

Average Score

56.7%

API Documentation Repository Model Weights Results Blog

Timeline

Key dates in the model's history

Announcement

August 25, 2026

Last Update

August 31, 2026

Technical Specifications

Parameters

8.0B

Training Tokens

Knowledge Cutoff

Family

Capabilities

MultimodalZeroEval

Benchmark Results

Model performance metrics across various tests and benchmarks

Programming

Programming skills tests

SWE-Bench Verified

Score • Self-reported

47.7%

Reasoning

Logical reasoning and analysis

GPQA

Score • Self-reported

64.1%

Other Tests

Specialized benchmarks

AIME 2025

Score • Self-reported

86.7%

Arena-Hard v2

Score • Self-reported

65.2%

BFCL-V4

Score • Self-reported

52.4%

Bird-SQL (dev)

Score • Self-reported

41.1%

HMMT25

Score • Self-reported

78.3%

IFBench

Score • Self-reported

79.3%

LiveCodeBench v6

Score • Self-reported

73.2%

MMLU-Pro

Score • Self-reported

74.0%

ProfBench

Score • Self-reported

41.2%

RULER 128k

Score • Self-reported

71.4%

RULER 64k

Score • Self-reported

81.0%

SciCode

Score • Self-reported

36.1%

SWE-bench Multilingual

Score • Self-reported

30.8%

SWE-Bench Pro

Score • Self-reported

19.1%

TAU3-Bench

Score • Self-reported

58.1%

Terminal-Bench 2.1

Score • Self-reported

20.6%

License & Metadata

License

apache_2_0

Announcement Date

August 25, 2026

Last Updated

August 31, 2026

Compare IBM Granite 4.2 8B

All comparisons

vs Hy3 vs Laguna S 2.1 vs Claude Opus 5 vs Hy4 preview vs GLM-5.3 Flash vs GPT-5.6 Terra vs Inkling-Small vs Claude Sonnet 5

Similar Models

All Models

IBM Granite 4.0 Tiny Preview

IBM

7.0B

Best score:0.8 (HumanEval)

Released:May 2025

Granite 3.3 8B Base

IBM

MM8.2B

Best score:0.9 (HumanEval)

Released:Apr 2025

Llama 3.1 8B Instruct

Phi-3.5-mini-instruct

Microsoft

3.8B

Best score:0.8 (ARC)

Released:Aug 2024

Price:$0.10/1M tokens

Llama 3.2 3B Instruct

Phi 4 Mini

Microsoft

3.8B

Best score:0.8 (ARC)

Released:Feb 2025

Gemma 2 9B

Google

9.2B

Best score:0.7 (MMLU)

Released:Jun 2024

Qwen2.5 7B Instruct

Alibaba

7.6B

Best score:0.8 (HumanEval)

Released:Sep 2024

Price:$0.30/1M tokens

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.