gpt-oss-120b

Composite indexes

• Artificial Analysis Intelligence Index#92 of 142
  12independentAA Intelligence Index v4.3.2, highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#60 of 88
  0%independentTerminal-Bench 4.0 (AA harness), highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#79 of 103
  26.2%independentTerminal-Bench 2.1 (AA harness), highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#83 of 94
  34%independenthighArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Claude Opus 5.5 at 66.9%

• LiveCodeBench#5 of 56
  87.8%independentAA, highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Fugu at 92.9%

Agents and tool use

• GDPval-AA#72 of 103
  596independentGDPval-AA, highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Claude Opus 5.5 at 1846

• τ²-Bench#39 of 88
  65.8%independenttelecom (AA), highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: GLM-5.2 at 99.1%

Reasoning and knowledge

• Humanity’s Last Exam#73 of 147
  19.6%independentno tools (AA), highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Claude Opus 5.5 at 61.4%

• GPQA Diamond#75 of 142
  78.2%independentAA, highArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: GPT-6 Astra at 96.1%

• CritPt#69 of 139
  1.1%independenthighArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: GPT-5.6 Sol at 32.3%

Vision and long context

• AA Long Context Reasoning#90 of 139
  52%independenthighArtificial Analysis: gpt-oss-120b (high) ↗

  Leader: Kimi K3 at 88.7%

Human preference

• LMArena Text#50 of 54
  1352independentgpt-oss-120b, style control default, vote cutoff 2026-09-13LMArena Text leaderboard ↗

  Leader: Claude Fable 5 at 1506

Other published results