Granite 4.2 30B

Composite indexes

• Artificial Analysis Intelligence Index#76 of 142
  15independentIntelligence Index v4.3; reasoning; estimatedArtificial Analysis: Granite 4.2 30B ↗

  Leader: Claude Opus 5.5 at 58

Coding

• SciCode#77 of 94
  38.8%vendorthinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗
  37.8%independentreasoningArtificial Analysis: Granite 4.2 30B ↗

  Leader: Claude Opus 5.5 at 66.9%

• Terminal-Bench 2.x#78 of 103
  29.2%vendorTerminal-Bench 2.1; thinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗
  26.6%independentTerminal-Bench 2.1; reasoningArtificial Analysis: Granite 4.2 30B ↗

  Leader: Claude Fable 5.1 at 91.4%

• SWE-bench Pro#30 of 31
  33.3%vendorthinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗

  Leader: Claude Opus 5.5 at 89.9%

• SWE-bench Verified#12 of 16
  57%vendorthinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗

  Leader: Claude Sonnet 5 at 85.2%

• LiveCodeBench#16 of 56
  75.8%vendorv6; thinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗

  Leader: Fugu at 92.9%

Agents and tool use

• GDPval-AA#73 of 103
  1225vendorGDPval (vendor-run)Hugging Face: ibm-granite/granite-4.2-30b model card ↗
  564independentreasoningArtificial Analysis: Granite 4.2 30B ↗

  Leader: Claude Opus 5.5 at 1846

Reasoning and knowledge

• GPQA Diamond#107 of 142
  66.4%vendorGPQA; thinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗
  64.4%independentreasoningArtificial Analysis: Granite 4.2 30B ↗

  Leader: GPT-6 Astra at 96.1%

• Humanity’s Last Exam#93 of 147
  11.2%independentno tools; reasoningArtificial Analysis: Granite 4.2 30B ↗

  Leader: Claude Opus 5.5 at 61.4%

• CritPt#85 of 139
  0.3%independentreasoningArtificial Analysis: Granite 4.2 30B ↗

  Leader: GPT-5.6 Sol at 32.3%

• HMMT February 2026#3 of 8
  89.2%vendorHMMT Feb 2025 (not 2026); thinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗

  Leader: GLM-5.2 at 92.5%

• MMLU-Pro#9 of 14
  77.6%vendorthinkingHugging Face: ibm-granite/granite-4.2-30b model card ↗

  Leader: Nemotron 3 Ultra 550B A55B at 86.8%

Vision and long context

• AA Long Context Reasoning#93 of 139
  49%independentreasoningArtificial Analysis: Granite 4.2 30B ↗

  Leader: Kimi K3 at 88.7%

Other published results