AAV Capsid Packaging Prediction
Can the model reason accurately about difficult scientific material?
Top rankings
One row per model, using its best reported score across effort settings.
Effort curve
Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.
Definition and comparison boundaryinternal methodology
Can the model reason accurately about difficult scientific material?
Scientific knowledge and reasoning under the benchmark's published question set. Higher is better. Spearman rank correlation reported by OpenAI.
This is a publisher-defined internal evaluation. The task set or grading details are not fully public, so treat it as directional evidence.
OpenAI reports Spearman correlation on 1,100 held-out AAV sequences after providing 5,000 labeled sequences and ESM embeddings; GPT-6 Astra is 0.528 and GPT-5.6 Sol is 0.529. The source's simple ESM-2 baseline is 0.288 and its indicative threshold is 0.600.
Method / source