scBench (BenchmarkList Legacy Multi-Harness): leaderboard
BenchmarkList imports single-cell RNA-seq agent workflow scores citing the legacy 195-task scBench release, alongside vendor system-card results. The imported rows include different agent harnesses whose exact dataset and run versions are not uniformly identified. This cohort remains separate from clean-v1 and the primary harness-specific results.
Metric: Accuracy (imported). Source: benchmarklist.com. Saturation forecast: Around October 2027. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Mythos 5 | 59.3 | #2 |
| 2 | Claude Opus 4.8 | 58.2 | #36 |
| 3 | Claude Mythos Preview | 58.2 | #3 |
| 4 | GPT-5.5 | 57.95 | #30 |
| 5 | GPT-5.4 | 57.44 | #84 |
| 6 | Claude Opus 4.7 | 55.3 | #51 |
| 7 | Gemini 3.1 Pro (Preview) | 53.85 | #59 |
| 8 | Claude Opus 4.6 | 52.65 | #64 |
| 9 | GPT-5.2 | 52.31 | #114 |
| 10 | Claude Sonnet 4.6 | 50.4 | #93 |
| 11 | Claude Opus 4.5 | 47.18 | #87 |
| 12 | Grok 4.20 Beta (0309) (Reasoning) | 44.44 | #210 (Grok 4.20 Beta (0309)) |
| 13 | Grok 4.3 | 44.27 | #192 |
| 14 | GPT-5.1 | 38.8 | #139 |
| 15 | Claude Sonnet 4.5 | 33.16 | #144 |
No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=scbench-benchmarklist-legacy-multi-harness · How It Works · Data refreshed daily, snapshot 2026-10-24.