GPT-5.6 Sol

Composite indexes

• Artificial Analysis Intelligence Index#8 of 142
  47independentAA Intelligence Index v4.3.2, max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#9 of 88
  39.9%independentTerminal-Bench 4.0 (AA harness), max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#8 of 103
  88%independentTerminal-Bench 2.1 (AA harness), max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#16 of 94
  57.1%independentmax effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Claude Opus 5.5 at 66.9%

Agents and tool use

• GDPval-AA#15 of 103
  1588independentGDPval-AA, max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Claude Opus 5.5 at 1846

• τ²-Bench#26 of 88
  85.1%independenttelecom (AA), max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: GLM-5.2 at 99.1%

• BrowseComp#3 of 14
  90.4%vendormax at any effort (from GPT-6 Astra comparison table)OpenAI: GPT-6 Astra ↗

  Leader: GPT-6 Astra at 91.5%

• OSWorld-Verified#13 of 16
  65.7%vendorOSWorld 2.0 offline (from GPT-6 Astra comparison table)OpenAI: GPT-6 Astra ↗

  Leader: Kimi K3 at 84.8%

Reasoning and knowledge

• Humanity’s Last Exam#7 of 147
  49.5%independentno tools (AA), max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Claude Opus 5.5 at 61.4%

• GPQA Diamond#8 of 142
  94.1%independentAA, max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: GPT-6 Astra at 96.1%

• CritPt#1 of 139
  32.3%independentmax effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: GPT-5.6 Sol at 32.3%

• ARC-AGI-2#3 of 25
  92.5%independentsemi-private, MaxARC Prize leaderboard ↗

  Leader: GPT-6 Astra at 95%

• ARC-AGI-3#4 of 9
  7.8%independentsemi-private, MaxARC Prize leaderboard ↗

  Leader: GPT-6 Astra at 62.7%

Vision and long context

• AA Long Context Reasoning#6 of 139
  84%independentmax effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Kimi K3 at 88.7%

• MMMU-Pro#7 of 72
  83.4%independentAA, max effortArtificial Analysis: GPT-5.6 Sol (max) ↗

  Leader: Claude Opus 5.5 at 87.7%

Human preference

• LMArena Text#11 of 54
  1483independentgpt-5.6-sol-xhigh, style control default, vote cutoff 2026-09-13LMArena Text leaderboard ↗

  Leader: Claude Fable 5 at 1506

• LMArena WebDev#17 of 54
  1617independentgpt-5.6-sol-xhigh (Codex harness), LMArena Code arena (formerly WebDev), vote cutoff 2026-09-23LMArena Code (WebDev) leaderboard ↗

  Leader: Claude Opus 5.5 at 1818

Other published results