1 benchmark · 34 results. Each benchmark shows the top of its default board (its first primary metric at its default version); ranks are the server's, ties share a rank, and "best" is the best visible result, not a state of the art.
Secondary structure (Q3) protein-ml/secondary-structure-q3
Q3 accuracy (per-residue probe) · ↑ higher is better · probe · atlas-n100-v2
| # | Model | Value [95% CI] | Source | Overlap |
|---|---|---|---|---|
| 1 | protein-ml/prot-t5-xl-half-uniref50-enc | 0.813 [0.799, 0.828] best | reproduced | overlapTrain/test overlap declared by the evaluator: BFD, protein-ml/uniref50 |
| 1 | protein-ml/prot-t5-xl-uniref50 | 0.813 [0.799, 0.828] best | reproduced | overlapTrain/test overlap declared by the evaluator: BFD, protein-ml/uniref50 |
| 3 | protein-ml/esm2-t30-150m-ur50d | 0.803 [0.787, 0.818] | reproduced | overlapTrain/test overlap declared by the evaluator: protein-ml/uniref50 |
| 4 | protein-ml/esm2-t33-650m-ur50d | 0.797 [0.783, 0.810] | reproduced | overlapTrain/test overlap declared by the evaluator: protein-ml/uniref50 |
| 5 | protein-ml/esm2-t36-3b-ur50d | 0.779 [0.765, 0.791] | reproduced | overlapTrain/test overlap declared by the evaluator: protein-ml/uniref50 |
5 of 17 on this board. Full leaderboard