Single Cell Analysis
Task 42 / 47
predict_modality
Predict surface-protein (ADT) modalities from single-cell RNA profiles in an Open Problems–style setup. Models must generalize under biological noise and are scored by benchmark-defined global and stratified metrics. It reflects multimodal integration in single-cell analysis.
Model leaderboard
Updated v1 results · 2026-09-15 · raw score, higher is better.
| # | Participant | Raw score | Medal |
|---|---|---|---|
| 1 | GPT-5.4 | 0.73096943 | Gold |
| 2 | Claude Opus 4.6 | 0.5467 | Silver |
| 2 | DeepSeek V3.2 | 0.5467 | Silver |
| 2 | Gemini 3.1 Pro Preview | 0.5467 | Silver |
| 2 | GLM-5 | 0.5467 | Silver |
| 2 | Grok 4.20 | 0.5467 | Silver |
| 2 | Qwen3 Coder Next | 0.5467 | Silver |
| 2 | Seed 2.0 Pro | 0.5467 | Silver |
Framework results · original paper
Historical results from the original evaluators · normalized score (0–100).
| # | Participant | Score |
|---|---|---|
| 1 | Claude Opus 4.6 + OpenEvolve | 100.0 |
| 2 | Claude Opus 4.6 + ShinkaiEvolve | 100.0 |
| 3 | GPT-OSS + ShinkaiEvolve | 48.0 |
| 4 | Claude Opus 4.6 + ABMCTS | 14.6 |
| 5 | GPT-OSS + ABMCTS | 0.2 |
| 6 | GPT-OSS + OpenEvolve | 0.0 |
Model results use the updated scores and medal thresholds. No valid score earns zero medal credit. Historical framework scores use the original evaluators and have not been updated alongside the model results.