Claude Sonnet 5

Composite indexes

• Artificial Analysis Intelligence Index#26 of 142
  38independentv4.3.2, max effortArtificial Analysis: Claude Sonnet 5 ↗
  32independentv4.3.2, high effort (API default)Artificial Analysis: Claude Sonnet 5 (high) ↗
  23independentv4.3.2, thinking offArtificial Analysis: Claude Sonnet 5 (non-reasoning) ↗

  Leader: Claude Opus 5.5 at 58

Coding

• SciCode#25 of 94
  54.3%independentAA run, max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: Claude Opus 5.5 at 66.9%

• Terminal-Bench 4.0#23 of 88
  14.1%independentAA run (Terminal-Bench 4.0), max effortArtificial Analysis: Claude Sonnet 5 ↗
  12.4%independentmax effort, Claude Code harness, ±3.1Terminal-Bench 4.0 official leaderboard ↗

  Leader: Claude Mythos 5.1 at 60.9%

• Terminal-Bench 2.x#22 of 103
  80.5%independentAA run (Terminal-Bench 2.1), max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: Claude Fable 5.1 at 91.4%

• SWE-bench Pro#8 of 31
  63.2%vendorAnthropic-run, max effort, avg of 5 trialsClaude Sonnet 5 System Card, Table 8.1.A ↗

  Leader: Claude Opus 5.5 at 89.9%

• SWE-bench Verified#1 of 16
  85.2%vendorAnthropic-run, max effort, avg of 5 trialsClaude Sonnet 5 System Card, section 8.2 ↗

  Leader: Claude Sonnet 5 at 85.2%

Agents and tool use

• GDPval-AA#22 of 103
  1449independentGDPval-AA v2.1, max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: Claude Opus 5.5 at 1846

• BrowseComp#5 of 14
  84.7%vendorsingle agent, max effort (multi-agent 86.6%)Claude Sonnet 5 System Card, Table 8.1.A ↗

  Leader: GPT-6 Astra at 91.5%

• OSWorld-Verified#5 of 16
  81.2%vendorOSWorld-Verified, max effortClaude Sonnet 5 System Card, Table 8.1.A ↗

  Leader: Kimi K3 at 84.8%

Reasoning and knowledge

• GPQA Diamond#32 of 142
  91.1%independentAA run, max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: GPT-6 Astra at 96.1%

• Humanity’s Last Exam#30 of 147
  57.4%vendorwith tools, max effortClaude Sonnet 5 System Card, Table 8.1.A ↗
  43.2%vendorno tools, max effortClaude Sonnet 5 System Card, Table 8.1.A ↗
  41.3%independentAA run, no tools, max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: Claude Opus 5.5 at 61.4%

• CritPt#27 of 139
  16.9%independentAA run, max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: GPT-5.6 Sol at 32.3%

Vision and long context

• AA Long Context Reasoning#18 of 139
  82%independentAA-LCR v1.1, max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: Kimi K3 at 88.7%

• MMMU-Pro#28 of 72
  77.3%independentAA run, max effortArtificial Analysis: Claude Sonnet 5 ↗

  Leader: Claude Opus 5.5 at 87.7%

Human preference

• LMArena Text#25 of 54
  1461independentclaude-sonnet-5-high; 35,301 votesLMArena text leaderboard ↗

  Leader: Claude Fable 5 at 1506

• LMArena WebDev#29 of 54
  1538independentclaude-sonnet-5-high; 11,645 votesLMArena Code Arena | WebDev leaderboard ↗

  Leader: Claude Opus 5.5 at 1818

Other published results