Claude Opus 5

Composite indexes

• Artificial Analysis Intelligence Index#4 of 142
  51independentv4.3.2, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Claude Opus 5.5 at 58

Coding

• SciCode#20 of 94
  56.4%independentAA run, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Claude Opus 5.5 at 66.9%

• Terminal-Bench 4.0#5 of 88
  53.9%independentxhigh effort, Claude Code harness, ±3.2Terminal-Bench 4.0 official leaderboard ↗
  51.8%independentmax effort, Claude Code harness, ±3.4Terminal-Bench 4.0 official leaderboard ↗
  49%independentAA run (Terminal-Bench 4.0), max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#5 of 103
  89.1%independentAA run (Terminal-Bench 2.1), max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Claude Fable 5.1 at 91.4%

• SWE-bench Pro#4 of 31
  79.2%vendorAnthropic-run, max effortClaude Opus 5.5 System Card, Table 8.1.A (Opus 5 column) ↗

  Leader: Claude Opus 5.5 at 89.9%

Agents and tool use

• GDPval-AA#3 of 103
  1708independentGDPval-AA v2.1, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Claude Opus 5.5 at 1846

• OSWorld-Verified#8 of 16
  74%vendorOSWorld 2.0 partial (strict 37.2%), revised harnessClaude Opus 5.5 System Card, Table 8.1.A (Opus 5 column) ↗

  Leader: Kimi K3 at 84.8%

Reasoning and knowledge

• GPQA Diamond#14 of 142
  93.2%independentAA run, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: GPT-6 Astra at 96.1%

• Humanity’s Last Exam#5 of 147
  63.6%vendorwith tools, max effortClaude Opus 5.5 System Card, Table 8.1.A (Opus 5 column) ↗
  56.6%vendorno tools, max effortClaude Opus 5.5 System Card, Table 8.1.A (Opus 5 column) ↗
  54.9%independentAA run, no tools, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Claude Opus 5.5 at 61.4%

• CritPt#7 of 139
  29.1%independentAA run, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: GPT-5.6 Sol at 32.3%

• ARC-AGI-2#4 of 25
  90.4%independentmax effort; ARC Prize verified, cost $2.06/taskARC Prize leaderboard data (semi-private) ↗

  Leader: GPT-6 Astra at 95%

• ARC-AGI-3#2 of 9
  30.2%independenthigh effort; ARC Prize semi-privateARC Prize leaderboard data (semi-private) ↗

  Leader: GPT-6 Astra at 62.7%

Vision and long context

• AA Long Context Reasoning#38 of 139
  79.3%independentAA-LCR v1.1, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Kimi K3 at 88.7%

• MMMU-Pro#5 of 72
  84.7%independentAA run, max effortArtificial Analysis: Claude Opus 5 ↗

  Leader: Claude Opus 5.5 at 87.7%

Human preference

• LMArena Text#7 of 54
  1493independentclaude-opus-5-high; 42,617 votesLMArena text leaderboard ↗

  Leader: Claude Fable 5 at 1506

• LMArena WebDev#4 of 54
  1692independentclaude-opus-5-max; 14,351 votesLMArena Code Arena | WebDev leaderboard ↗

  Leader: Claude Opus 5.5 at 1818

Other published results