IBM Granite 4.2 30B
IBM Granite 4.2 30B is the flagship 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. The 8B and 30B sizes received additional agentic reinforcement learning for multi-step tool use. Open weights on Hugging Face under Apache 2.0. Text-only (not multimodal).
Key Specifications
Parameters
30.0B
Context
-
Release Date
August 25, 2026
Average Score
62.4%
Timeline
Key dates in the model's history
Announcement
August 25, 2026
Last Update
September 4, 2026
Today
September 5, 2026
Technical Specifications
Parameters
30.0B
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
Reasoning
Logical reasoning and analysis
GPQA
Score • Self-reported
Other Tests
Specialized benchmarks
AIME 2025
Score • Self-reported
Arena-Hard v2
Score • Self-reported
BFCL-V4
Score • Self-reported
Bird-SQL (dev)
Score • Self-reported
HMMT25
Score • Self-reported
IFBench
Score • Self-reported
LiveCodeBench v6
Score • Self-reported
MMLU-Pro
Score • Self-reported
ProfBench
Score • Self-reported
RULER 128k
Score • Self-reported
RULER 64k
Score • Self-reported
SciCode
Score • Self-reported
SWE-bench Multilingual
Score • Self-reported
SWE-Bench Pro
Score • Self-reported
TAU3-Bench
Score • Self-reported
Terminal-Bench 2.1
Score • Self-reported
License & Metadata
License
apache_2_0
Announcement Date
August 25, 2026
Last Updated
September 4, 2026
Compare IBM Granite 4.2 30B
All comparisonsSimilar Models
All ModelsIBM Granite 4.0 Tiny Preview
IBM
7.0B
Best score:0.8 (HumanEval)
Released:May 2025
IBM Granite 4.2 8B
IBM
8.0B
Best score:0.6 (GPQA)
Released:Aug 2026
Jamba 1.5 Mini
AI21 Labs
52.0B
Best score:0.9 (ARC)
Released:Aug 2024
Price:$0.20/1M tokens
Phi-3.5-MoE-instruct
Microsoft
60.0B
Best score:0.9 (ARC)
Released:Aug 2024
Llama-3.3 Nemotron Super 49B v1
NVIDIA
49.9B
Best score:0.7 (GPQA)
Released:Mar 2025
DeepSeek R1 Distill Llama 70B
DeepSeek
70.6B
Best score:0.7 (GPQA)
Released:Jan 2025
Price:$0.10/1M tokens
DeepSeek R1 Distill Qwen 32B
DeepSeek
32.8B
Best score:0.6 (GPQA)
Released:Jan 2025
Price:$0.12/1M tokens
Gemma 2 27B
27.2B
Best score:0.8 (MMLU)
Released:Jun 2024
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.