Flagship successor to V3 (Max Effort reasoning). Open-weight, competitive with closed frontier at a fraction of the cost. Awaiting Multimodal eval coverage.
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 83 models released within six months of it, across 5 benchmarks where scores spread out.
| Benchmark | Score | Rank |
|---|---|---|
Arena EloArtificial Analysis Human preference ranking via blind comparisons | 1553 | #4 / 51 |
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 87.5% | #13 / 62 |
GPQAArtificial Analysis PhD-level science questions even experts struggle with | 92.8% | #19 / 96 |
TerminalArtificial Analysis Agentic terminal coding tasks requiring multi-step execution | 46.2% | #20 / 58 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 87.2% | #31 / 61 |
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 41% | #40 / 102 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisDeepSeek V4 Pro 0813 (Max) | 41% |
Terminal-Bench 2.1 Artificial AnalysisDeepSeek V4 Pro 0813 (Max) | 78.7% |
Terminal-Bench 4.0 Artificial AnalysisDeepSeek V4 Pro 0813 (Max) | 14.1% |
SciCode Artificial AnalysisDeepSeek V4 Pro 0813 (Max) | 51% |
AA-LCR Artificial AnalysisDeepSeek V4 Pro 0813 (Max) | 80.3% |
CritPt Artificial AnalysisDeepSeek V4 Pro 0813 (Max) | 18% |