1M context, stronger coding and vision. SWE-bench Pro 64.3% beats GPT-5.4 and Gemini 3.1 Pro.
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 82 models released within six months of it, across 7 benchmarks where scores spread out.
| Benchmark | Score | Rank |
|---|---|---|
SWE-bench Real-world GitHub issue resolution | 87.6% | #1 / 39 |
Arena Elo Human preference ranking via blind comparisons | 1753 | #1 / 51 |
OSWorld Computer use in real desktop environments | 78% | #2 / 13 |
GPQA PhD-level science questions even experts struggle with | 94.2% | #6 / 96 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 89.9% | #8 / 61 |
Terminal Agentic terminal coding tasks requiring multi-step execution | 69.4% | #10 / 58 |
ARC-AGIARC Prize Novel reasoning tasks requiring fluid intelligence | 82.2% | #12 / 40 |
MMLU Tests knowledge across 57 subjects from STEM to humanities | 91.5% | #14 / 54 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 85.5% | #23 / 51 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 84.7% | #30 / 62 |
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 42.3% | #34 / 100 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisClaude Opus 4.7 (Max) | 42.3% |
MMMU-Pro Artificial AnalysisClaude Opus 4.7 (Max) | 78.8% |
Terminal-Bench 2.1 Artificial AnalysisClaude Opus 4.7 (Max) | 83.1% |
AA-LCR Artificial AnalysisClaude Opus 4.7 (Max) | 78.7% |
CritPt Artificial AnalysisClaude Opus 4.7 (Max) | 12% |
ITBench SRE Artificial AnalysisClaude Opus 4.7 (Max) | 46.7% |
IFBench Artificial AnalysisClaude Opus 4.7 (Max) | 58.6% |