Claude 5.5 model for agentic coding and knowledge work. Independent measurements track max effort with default safeguards and fallback behavior.
SI v3 uses weighted percentages from one Artificial Analysis configuration. Ranking requires all five benchmarks. Partial scores average only available weights. Scores are not comparable to SI v2 or a probability of reaching singularity.
Compared with 58 models released within six months of it, across 4 benchmarks where scores spread out.
| Benchmark | Score | Rank |
|---|---|---|
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 61.4% | #1 / 95 |
ARC-AGIARC Prize Novel reasoning tasks requiring fluid intelligence | 97.9% | #2 / 39 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 61.4% |
MMMU-Pro Artificial AnalysisClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 87.7% |
Terminal-Bench 4.0 Artificial AnalysisClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 59.6% |
SciCode Artificial AnalysisClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 66.9% |
AA-LCR Artificial AnalysisClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 84.7% |
CritPt Artificial AnalysisClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 31.7% |
ITBench SRE Artificial AnalysisClaude Opus 5.5 (Adaptive Reasoning, Max Effort, Default Fallback) | 38.2% |
Terminal-Bench 4.0 Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks; Xhigh effort; standard error 2.6 points | 66.4% |
FrontierCode 1.1 Main Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks | 54.4% |
CursorBench 4.0 Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks | 57.8% |
GDPval-AA 2.1 Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks | 1846 Elo |
AutomationBench Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks; Zapier early-access evaluation; no fallback models | 40% |
Humanity's Last Exam Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks; With tools | 67.7% |
Terminal-Bench-Science 0.1 Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks | 58.7% |
OSWorld 2.0 Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks; Partial reward | 81.8% |
Chartography Provider reportAdaptive thinking, max effort unless noted; production safeguards with Opus 4.8/5 fallbacks; With tools | 89% |