Mid-tier model between Luna and Sol, built for everyday computer use across apps. OpenAI says it finishes agent tasks about three times faster than GPT-6.1 Sol.
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 57 models released within six months of it, across 3 benchmarks where scores spread out.
| Benchmark | Score | Rank |
|---|---|---|
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 46.2% | #22 / 102 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisGPT-6.1 Terra (Max) | 46.2% |
MMMU-Pro Artificial AnalysisGPT-6.1 Terra (Max) | 84.7% |
Terminal-Bench 4.0 Artificial AnalysisGPT-6.1 Terra (Max) | 55.6% |
SciCode Artificial AnalysisGPT-6.1 Terra (Max) | 55.1% |
AA-LCR Artificial AnalysisGPT-6.1 Terra (Max) | 82.7% |
CritPt Artificial AnalysisGPT-6.1 Terra (Max) | 28% |