gpt-oss-20b

Composite indexes

• Artificial Analysis Intelligence Index#114 of 151
  9independentAA Intelligence Index v4.3.2, highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#69 of 96
  0%independentTerminal-Bench 4.0 (AA harness), highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Claude Sonnet 5.5 at 70.6%

• Terminal-Bench 2.x#89 of 105
  13.9%independentTerminal-Bench 2.1 (AA harness), highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#78 of 100
  38.9%independenthighArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Claude Opus 5.5 at 66.9%

• LiveCodeBench#12 of 57
  77.7%independentAA, highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Fugu at 92.9%

Agents and tool use

• GDPval-AA#91 of 109
  327independentGDPval-AA, highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Muse Wonder Full at 1857

• τ²-Bench#41 of 89
  60.2%independenttelecom (AA), highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: GLM-5.2 at 99.1%

Reasoning and knowledge

• Humanity’s Last Exam#101 of 153
  11%independentno tools (AA), highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Claude Sonnet 5.5 at 64.5%

• GPQA Diamond#102 of 144
  68.8%independentAA, highArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: GPT-6 Astra at 96.1%

• CritPt#71 of 144
  1.4%independenthighArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Muse Wonder Full at 33%

Vision and long context

• AA Long Context Reasoning#107 of 144
  34.7%independenthighArtificial Analysis: gpt-oss-20b (high) ↗

  Leader: Kimi K3 at 88.7%

Human preference

• LMArena Text#57 of 57
  1317independentgpt-oss-20b, style control default, vote cutoff 2026-09-13LMArena Text leaderboard ↗

  Leader: Gemini 4 Argon at 1525

Other published results