Apodex Discovery - AAV Viability Prediction: leaderboard
Apodex Discovery (2026) puts agentic solvers in a domain-specific environment for adeno-associated virus capsid design with real screening data and hidden verifiers. This task asks the solver to predict whether capsid variants stay viable, scored by out-of-distribution AUROC on a 0 to 1 scale; higher is better.
Metric: Out-of-distribution AUROC (0 to 1) for predicting whether adeno-associated virus capsid variants remain viable, separating viable from non-viable variants with the same number of mutations, averaged over a multi-mutation and a single-mutation instance; solver runs in the authors' domain-specific environment. Source: arxiv.org. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Kimi K3 | 0.94 | #53 |
| 2 | Claude Opus 5 | 0.94 | #20 |
| 3 | GLM-5.2 | 0.93 | #137 |
| 4 | Claude Opus 4.8 | 0.93 | #56 |
| 5 | Claude Sonnet 4.6 | 0.92 | #158 |
| 6 | DeepSeek V4 Pro | 0.92 | #181 |
| 7 | GPT-5.5 | 0.91 | #48 |
| 8 | Apodex 1.1 | 0.9 | #282 |
| 9 | Qwen 3.5 397B A17B | 0.9 | #242 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=apodex-discovery-aav-viability-prediction · How It Works · Data refreshed daily, snapshot 2026-11-12.