First model to break 1500 Arena Elo, 100% on AIME with code execution
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 38 models released within six months of it, across 9 benchmarks where scores spread out.
| Benchmark | Score | Rank |
|---|---|---|
MMLU Tests knowledge across 57 subjects from STEM to humanities | 93.2% | #2 / 54 |
ARC-C Grade-school science questions requiring reasoning | 98.8% | #2 / 40 |
HellaSwag Common sense reasoning about everyday situations | 97.9% | #2 / 36 |
FrontierMath Unpublished research-level mathematics problems | 38% | #3 / 4 |
HumanEval Coding ability - generating correct Python functions | 96.2% | #4 / 50 |
MATH Competition-level mathematics problems | 100% | #4 / 50 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 90.1% | #6 / 61 |
Arena Elo Human preference ranking via blind comparisons | 1501 | #10 / 51 |
SWE-bench Real-world GitHub issue resolution | 76.2% | #14 / 39 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 87.5% | #14 / 51 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 86.4% | #20 / 62 |
ARC-AGI Novel reasoning tasks requiring fluid intelligence | 45.1% | #27 / 40 |
GPQA PhD-level science questions even experts struggle with | 91.9% | #28 / 96 |
TerminalArtificial Analysis Agentic terminal coding tasks requiring multi-step execution | 41.7% | #29 / 58 |
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 39.7% | #44 / 100 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisGemini 3 Pro Preview (High) | 39.7% |
MMMU-Pro Artificial AnalysisGemini 3 Pro Preview (High) | 80.2% |
AA-LCR Artificial AnalysisGemini 3 Pro Preview (High) | 76% |
CritPt Artificial AnalysisGemini 3 Pro Preview (High) | 9.1% |
IFBench Artificial AnalysisGemini 3 Pro Preview (High) | 70.4% |