Agentic coding focus, 1M context, strong multimodal performance
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 78 models released within six months of it, across 5 benchmarks where scores spread out.
| Benchmark | Score | Rank |
|---|---|---|
SWE-bench Real-world GitHub issue resolution | 78.8% | #8 / 39 |
Terminal Agentic terminal coding tasks requiring multi-step execution | 61.6% | #13 / 58 |
Arena Elo Human preference ranking via blind comparisons | 1470 | #17 / 51 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 86% | #21 / 62 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 87.7% | #24 / 61 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 84.2% | #25 / 51 |
GPQAArtificial Analysis PhD-level science questions even experts struggle with | 88.2% | #51 / 96 |
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 27.8% | #71 / 100 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisQwen3.6 Plus | 27.8% |
MMMU-Pro Artificial AnalysisQwen3.6 Plus | 78% |
Terminal-Bench 2.1 Artificial AnalysisQwen3.6 Plus | 61.4% |
AA-LCR Artificial AnalysisQwen3.6 Plus | 78.3% |
CritPt Artificial AnalysisQwen3.6 Plus | 2.9% |
IFBench Artificial AnalysisQwen3.6 Plus | 75.2% |