First public Mythos-class model, a tier above Opus, with strong reasoning, coding, and long-horizon professional-work 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 79 models released within six months of it, across 4 benchmarks where scores spread out.
| Benchmark | Score | Rank |
|---|---|---|
LiveCodeBenchvals.ai Contamination-free competitive programming (filtered by cutoff date) | 89.8% | #2 / 62 |
MMLU-Provals.ai Harder 10-option successor to MMLU; more reasoning-focused | 91.5% | #3 / 61 |
MMMUvals.ai College-level multimodal reasoning across 30+ disciplines | 89.3% | #3 / 51 |
AA Intelligence (Historical) Archived release-era composite, not comparable to the current versioned Intelligence Index used for rankings | 59.9% | #4 / 22 |
AA Coding (Historical) Archived release-era coding composite, excluded from current rankings because comparable current-series coverage is unavailable | 77.2% | #4 / 22 |
HLEArtificial Analysis Challenging multidisciplinary questions evaluated by Artificial Analysis | 55.5% | #4 / 102 |
Terminal Agentic terminal coding tasks requiring multi-step execution | 83.1% | #5 / 58 |
GPQAArtificial Analysis PhD-level science questions even experts struggle with | 92.6% | #22 / 96 |
| Benchmark / source | Score |
|---|---|
HLE Artificial AnalysisClaude Fable 5 (Max, Opus 4.8 Fallback) | 55.5% |
Terminal-Bench 2.1 Artificial AnalysisClaude Fable 5 (Max, Opus 4.8 Fallback) | 84.6% |
Terminal-Bench 4.0 Artificial AnalysisClaude Fable 5 (Max, Opus 4.8 Fallback) | 42.4% |
SciCode Artificial AnalysisClaude Fable 5 (Max, Opus 4.8 Fallback) | 61% |
AA-LCR Artificial AnalysisClaude Fable 5 (Max, Opus 4.8 Fallback) | 82.3% |
CritPt Artificial AnalysisClaude Fable 5 (Max, Opus 4.8 Fallback) | 28.6% |
IFBench Artificial AnalysisClaude Fable 5 (Max, Opus 4.8 Fallback) | 63.5% |