Model type
Multitrack generative protein transformer
Model
ESM3
ESM3 generates and completes protein sequence, structure and functional annotations using a shared multimodal representation.
Sources (3)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and LicensesHow it worksESM3 workflow
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Sources (3)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and LicensesOverview
Inputs
Complete or partial protein sequence, structure and function-keyword tracks.
Outputs
Completed sequence, structure and functional tracks.
Access
Official project documentation and implementation: https://github.com/evolutionaryscale/esm
Sources (3)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenseslimited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
60 evaluations · 77 results. Different protocols are not a single leaderboard.
Sorted by Plddt (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.
| Tested configuration | Protocol and dataset | Plddt | Sctm | Proteinbench max clust | Proteinbench max TM | Spearman | Accuracy | Perplexity | Proteinbench pairwise TM | Scrmsd | AUROC | F1 max (multi-label; TorchDrug-style F1_score_max) | Spearman (ProteinGym zero-shot DMS substitutions, Activity assays: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Binding assays: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Expression assays: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, High MSA depth proteins: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Human proteins: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Low MSA depth proteins: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, mean over selection types of the per-type mean over proteins of assay-level Spearman (all 217 assays)) | Spearman (ProteinGym zero-shot DMS substitutions, Medium MSA depth proteins: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Organismal fitness assays: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Other eukaryote proteins: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Prokaryote proteins: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Stability assays: mean over proteins of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, variants with 1 substitution: mean over protein and selection-type groups of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, variants with 2 substitutions: mean over protein and selection-type groups of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, variants with 3 substitutions: mean over protein and selection-type groups of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, variants with 4 substitutions: mean over protein and selection-type groups of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, variants with 5 or more substitutions: mean over protein and selection-type groups of assay-level Spearman) | Spearman (ProteinGym zero-shot DMS substitutions, Virus proteins: mean over proteins of assay-level Spearman) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-100-PLDDT: De novo backbones based sequence design, length 100 pLDDT, CASP, CAMEO and de novo backbones (ProteinBench split) | 86.6 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-300-PLDDT: De novo backbones based sequence design, length 300 pLDDT, CASP, CAMEO and de novo backbones (ProteinBench split) | 70.8 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-400-PLDDT: De novo backbones based sequence design, length 400 pLDDT, CASP, CAMEO and de novo backbones (ProteinBench split) | 62.6 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-200-PLDDT: De novo backbones based sequence design, length 200 pLDDT, CASP, CAMEO and de novo backbones (ProteinBench split) | 60.7 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-500-PLDDT: De novo backbones based sequence design, length 500 pLDDT, CASP, CAMEO and de novo backbones (ProteinBench split) | 59.4 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-200-QUALITY-PLDDT: length 200, Quality pLDDT, Generated sequences at fixed lengths (ProteinBench split) | 58.45 ± 9.40 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-100-QUALITY-PLDDT: length 100, Quality pLDDT, Generated sequences at fixed lengths (ProteinBench split) | 54.26 ± 15.35 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-500-QUALITY-PLDDT: length 500, Quality pLDDT, Generated sequences at fixed lengths (ProteinBench split) | 52.17 ± 10.52 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-300-QUALITY-PLDDT: length 300, Quality pLDDT, Generated sequences at fixed lengths (ProteinBench split) | 48.08 ± 13.34 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench ANTI-RES: Antibiotic resistance, Antibiotic resistance (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | 0.584 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench BINDING-DB: BindingDB, BindingDB (PFMBench split) | Not reported | Not reported | Not reported | Not reported | 0.225 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench CLONING-CLF: Cloning CLF, Cloning CLF (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.774 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench DEEPLOC2: DeepLoc2 Multi, DeepLoc2 Multi (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.659 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench DEEPSOL: DeepSol, DeepSol (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.781 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench EC: Enzyme Commission, Enzyme Commission (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.648 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench MAT-PROD: Material production, Material production (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | 0.775 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench METAL-ION: Metal ion binding, Metal ion binding (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | 0.703 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench PDB-BIND: PDBbind, PDBbind (PFMBench split) | Not reported | Not reported | Not reported | Not reported | 0.156 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench PROTEINGYM-ZS: ProteinGym zero-shot variant effect prediction, ProteinGym (PFMBench split) | Not reported | Not reported | Not reported | Not reported | 0.414 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench SEC-STRUCT: Secondary structure, Secondary structure (PFMBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | 0.813 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | PFMBench STABILITY: TAPE_Stability, TAPE_Stability (PFMBench split) | Not reported | Not reported | Not reported | Not reported | 0.157 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust., Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 0.9 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-100-DIVERSITY-PAIRWISE-TM: length 100, Diversity pairwise TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.45 ± 0.15 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.68 ± 0.07 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-100-QUALITY-PPL: length 100, Quality ppl, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 14.79 ± 2.90 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust., Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 1 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-200-DIVERSITY-PAIRWISE-TM: length 200, Diversity pairwise TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.35 ± 0.07 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.80 ± 0.01 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-200-QUALITY-PPL: length 200, Quality ppl, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 12.96 ± 2.38 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust., Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 1 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-300-DIVERSITY-PAIRWISE-TM: length 300, Diversity pairwise TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.32 ± 0.03 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.75 ± 0.02 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-300-QUALITY-PPL: length 300, Quality ppl, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 14.59 ± 2.97 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-500-DIVERSITY-MAX-CLUST: length 500, Diversity Max Clust., Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 1 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-500-DIVERSITY-PAIRWISE-TM: length 500, Diversity pairwise TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.30 ± 0.05 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-500-NOVELTY-MAX-TM: length 500, Novelty Max TM, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.54 ± 0.03 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 (1.4B)Evaluation | ProteinBench SEQ-LENGTH-500-QUALITY-PPL: length 500, Quality ppl, Generated sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 11.10 ± 2.26 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust., Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 0.64 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.41 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-100-QUALITY-SCRMSD: length 100, Quality scRMSD, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 13.80 ± 10.51 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-100-QUALITY-SCTM: length 100, Quality scTM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | 0.72 ± 0.19 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust., Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 0.63 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.61 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-200-QUALITY-SCRMSD: length 200, Quality scRMSD, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 21.18 ± 16.19 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-200-QUALITY-SCTM: length 200, Quality scTM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | 0.63 ± 0.20 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust., Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 0.52 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.73 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-300-QUALITY-SCRMSD: length 300, Quality scRMSD, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 25.5 ± 20.68 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-300-QUALITY-SCTM: length 300, Quality scTM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | 0.59 ± 0.21 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-500-DIVERSITY-MAX-CLUST: length 500, Diversity Max Clust., Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | 0.46 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-500-NOVELTY-MAX-TM: length 500, Novelty Max TM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | 0.78 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-500-QUALITY-SCRMSD: length 500, Quality scRMSD, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 26.72 ± 21.08 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench CO-LENGTH-500-QUALITY-SCTM: length 500, Quality scTM, Co-generated structures and sequences at fixed lengths (ProteinBench split) | Not reported | 0.64 ± 0.20 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-100-SCTM: De novo backbones based sequence design, length 100 scTM, CASP, CAMEO and de novo backbones (ProteinBench split) | Not reported | 0.942 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-200-SCTM: De novo backbones based sequence design, length 200 scTM, CASP, CAMEO and de novo backbones (ProteinBench split) | Not reported | 0.486 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-300-SCTM: De novo backbones based sequence design, length 300 scTM, CASP, CAMEO and de novo backbones (ProteinBench split) | Not reported | 0.632 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-400-SCTM: De novo backbones based sequence design, length 400 scTM, CASP, CAMEO and de novo backbones (ProteinBench split) | Not reported | 0.564 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3Evaluation | ProteinBench IF-DE-NOVO-BACKBONES-BASED-SEQUENCE-DESIGN-LENGTH-500-SCTM: De novo backbones based sequence design, length 500 scTM, CASP, CAMEO and de novo backbones (ProteinBench split) | Not reported | 0.452 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 open (1.4B) [ProteinGym 144fe22b snapshot]Evaluation | ProteinGym AMFR official zero-shot Spearman snapshot, ProteinGym AMFR 4G3O official snapshot cohort | Not reported | Not reported | Not reported | Not reported | 0.29 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| ESM3 open (1.4B) [ProteinGym 144fe22b snapshot]Evaluation | ProteinGym v1.3 zero-shot DMS substitutions, ProteinGym DMS substitution assays at commit 144fe22 (217 assays) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 0.43 | 0.4 | 0.47 | 0.575 | 0.48 | 0.397 | 0.466 | 0.466 | 0.389 | 0.548 | 0.53 | 0.641 | 0.487 | 0.337 | 0.307 | 0.285 | 0.363 | 0.406 |
All 77 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: ESM3 | Task: PFMBench ANTI-RES: Antibiotic resistance Dataset subset: Antibiotic resistance (PFMBench split) | 0.584 Accuracy fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench ANTI-RES: Antibiotic resistance Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(Anti.Res.) |
| Configuration: ESM3 | Task: PFMBench BINDING-DB: BindingDB Dataset subset: BindingDB (PFMBench split) | 0.225 Spearman unitless · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench BINDING-DB: BindingDB Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(Bind. DB) |
| Configuration: ESM3 | Task: PFMBench CLONING-CLF: Cloning CLF Dataset subset: Cloning CLF (PFMBench split) | 0.774 AUROC unitless · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench CLONING-CLF: Cloning CLF Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(Clo. CLF) |
| Configuration: ESM3 | Task: PFMBench DEEPLOC2: DeepLoc2 Multi Dataset subset: DeepLoc2 Multi (PFMBench split) | 0.659 F1 max (multi-label; TorchDrug-style F1_score_max) fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench DEEPLOC2: DeepLoc2 Multi Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(DL2 M.) |
| Configuration: ESM3 | Task: PFMBench DEEPSOL: DeepSol Dataset subset: DeepSol (PFMBench split) | 0.781 AUROC unitless · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench DEEPSOL: DeepSol Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(DeepSol) |
| Configuration: ESM3 | Task: PFMBench EC: Enzyme Commission Dataset subset: Enzyme Commission (PFMBench split) | 0.648 F1 max (multi-label; TorchDrug-style F1_score_max) fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench EC: Enzyme Commission Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(EC) |
| Configuration: ESM3 | Task: PFMBench MAT-PROD: Material production Dataset subset: Material production (PFMBench split) | 0.775 Accuracy fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench MAT-PROD: Material production Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(Mat.Pro.) |
| Configuration: ESM3 | Task: PFMBench METAL-ION: Metal ion binding Dataset subset: Metal ion binding (PFMBench split) | 0.703 Accuracy fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench METAL-ION: Metal ion binding Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(M. I. Bin.) |
| Configuration: ESM3 | Task: PFMBench PDB-BIND: PDBbind Dataset subset: PDBbind (PFMBench split) | 0.156 Spearman unitless · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench PDB-BIND: PDBbind Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(PDBBind) |
| Configuration: ESM3 | Task: PFMBench PROTEINGYM-ZS: ProteinGym zero-shot variant effect prediction Dataset subset: ProteinGym (PFMBench split) | 0.414 Spearman unitless · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench PROTEINGYM-ZS: ProteinGym zero-shot variant effect prediction Scored zero-shot, with no fine-tuning, as reported in PFMBench Table 4. Aggregation: Not reported pfmbench primary benchmark evidence · Table 4, row(ESM3 [ 18 ]), column(ProteinGym) |
| Configuration: ESM3 | Task: PFMBench SEC-STRUCT: Secondary structure Dataset subset: Secondary structure (PFMBench split) | 0.813 Accuracy fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench SEC-STRUCT: Secondary structure Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(Sec. Str.) |
| Configuration: ESM3 | Task: PFMBench STABILITY: TAPE_Stability Dataset subset: TAPE_Stability (PFMBench split) | 0.157 Spearman unitless · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 on PFMBench STABILITY: TAPE_Stability Fine-tuned with an adapter under the PFMBench harness; train, validation and test counts are in Table 1. Aggregation: Not reported pfmbench primary benchmark evidence · Table 3, row(ESM3 [ 18 ]), column(Stability) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust. Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 0.9 Proteinbench max clust fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust. Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 100, Diversity Max Clust. ↑) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-100-DIVERSITY-PAIRWISE-TM: length 100, Diversity pairwise TM Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 0.45 ± 0.15 Proteinbench pairwise TM unitless · lower Uncertainty: SD 0.15 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-100-DIVERSITY-PAIRWISE-TM: length 100, Diversity pairwise TM Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 100, Diversity pairwise TM ↓) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TM Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 0.68 ± 0.07 Proteinbench max TM unitless · lower Uncertainty: SD 0.07 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TM Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 100, Novelty Max TM ↓) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-100-QUALITY-PLDDT: length 100, Quality pLDDT Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 54.26 ± 15.35 Plddt unitless · higher Uncertainty: SD 15.35 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-100-QUALITY-PLDDT: length 100, Quality pLDDT Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 100, Quality pLDDT ↑) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-100-QUALITY-PPL: length 100, Quality ppl Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 14.79 ± 2.90 Perplexity unitless · lower Uncertainty: SD 2.90 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-100-QUALITY-PPL: length 100, Quality ppl Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 100, Quality ppl ↓) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust. Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 1 Proteinbench max clust fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust. Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 200, Diversity Max Clust. ↑) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-200-DIVERSITY-PAIRWISE-TM: length 200, Diversity pairwise TM Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 0.35 ± 0.07 Proteinbench pairwise TM unitless · lower Uncertainty: SD 0.07 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-200-DIVERSITY-PAIRWISE-TM: length 200, Diversity pairwise TM Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 200, Diversity pairwise TM ↓) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TM Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 0.80 ± 0.01 Proteinbench max TM unitless · lower Uncertainty: SD 0.01 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TM Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 200, Novelty Max TM ↓) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-200-QUALITY-PLDDT: length 200, Quality pLDDT Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 58.45 ± 9.40 Plddt unitless · higher Uncertainty: SD 9.40 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-200-QUALITY-PLDDT: length 200, Quality pLDDT Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 200, Quality pLDDT↑) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-200-QUALITY-PPL: length 200, Quality ppl Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 12.96 ± 2.38 Perplexity unitless · lower Uncertainty: SD 2.38 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-200-QUALITY-PPL: length 200, Quality ppl Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 200, Quality ppl ↓) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust. Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 1 Proteinbench max clust fraction · higher Uncertainty: Not yet extracted Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust. Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 300, Diversity Max Clust. ↑) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-300-DIVERSITY-PAIRWISE-TM: length 300, Diversity pairwise TM Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 0.32 ± 0.03 Proteinbench pairwise TM unitless · lower Uncertainty: SD 0.03 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-300-DIVERSITY-PAIRWISE-TM: length 300, Diversity pairwise TM Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 300, Diversity pairwise TM ↓) |
| Configuration: ESM3 (1.4B) | Task: ProteinBench SEQ-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TM Dataset subset: Generated sequences at fixed lengths (ProteinBench split) | 0.75 ± 0.02 Proteinbench max TM unitless · lower Uncertainty: SD 0.02 Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESM3 (1.4B) on ProteinBench SEQ-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TM Unconditional sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Aggregation: Not reported proteinbench primary benchmark evidence · Table 4, row(ESM3 (1.4B)), column(length 300, Novelty Max TM ↓) |
Source checking is not independent reproduction. Release 2026-10-22-1b7cf027e4f6.
Related configurations, pipelines and services
These configurations, services and pipelines use this model within their own configurations. Their results, where available, are not assigned to the underlying model.
- MegSite + ESM3 Pipeline · 1 result
Use this model
How it works, versions and access
Versions and evaluated configurations
- ESM3 · family
- ESM3 · family
- ESM3 (1.4B) · family
- ESM3 open (1.4B) [ProteinGym 144fe22b snapshot] · configuration of
How it works
How it works
Transformer generative masked-language model with discrete sequence, structure and function tracks; generation iteratively fills masked positions. The documented inputs are complete or partial protein sequence, structure and function-keyword tracks. The output consists of completed sequence, structure and functional tracks.
Sources (3)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and LicensesVersions and reproducibility
Published March 2024 small/medium/large models, August 2024 small/medium models, and esm3-sm-open-v1 local weights. The applicable input limits require configuration-specific checking.
Sources (3)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and LicensesStrengths, limitations and unresolved questions
Strengths and limitations
Strengths and considerations
- Partial prompts can constrain more than one biological property in the same generation workflow.
Sources (3)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses
Limitations and conditions
- Open local weights and API model names refer to different released sizes and dates; results require the actual configuration, not only the ESM3 family name.
Sources (3)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses
Profile review details
Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.
Stable record: discovery-model-esm3Specifications
Inputs, training, access and other details
Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.
| Property | Description and evidence |
|---|---|
| Model type | Multitrack generative protein transformerSources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Architecture | Transformer generative masked-language model with discrete sequence, structure and function tracks; generation iteratively fills masked positions.Sources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Inputs | Complete or partial protein sequence, structure and function-keyword tracks.Sources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Outputs | Completed sequence, structure and functional tracks.Sources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Parameters | 1.4B small, 7B medium and 98B large variants.Sources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Known versions | Published March 2024 small/medium/large models, August 2024 small/medium models, and esm3-sm-open-v1 local weights.Sources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Training data | ESM3 overview reports 2.78 billion proteins and 771 billion unique tokens for the largest model; this is not a verified per-checkpoint manifest.Sources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Training cutoff | The inspected ESM3 release documentation does not supply one latest-data date shared across its sequence, structure and function tracks. · Not reported in inspected sourcesSources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Context limits | The inspected ESM3 family documentation does not establish one context limit for local small, hosted medium and hosted large models; a specific service/checkpoint is required. · Not reported in inspected sourcesSources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Weights licence | MIT stated in the inspected ESM3 README; model and API access conditions remain separate.Sources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Access | Official project documentation and implementation: https://github.com/evolutionaryscale/esmSources (3)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses |
| Code licence | MITSourcesevolutionaryscale/esm: LICENSE.md · LICENSE.md: licence text |
Applicable tests and references
Applicability is distinct from a completed evaluation.
- ProteinBench · Proposed association
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
58 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation. Individual claims | evolutionaryscale/esm: _assets/ESM3_README.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram caption Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation. Individual claims | evolutionaryscale/esm: THIRD_PARTY_NOTICE.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram caption Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation. Individual claims | evolutionaryscale/esm: README.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
Diagram steps
| evolutionaryscale/esm: _assets/ESM3_README.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
Diagram steps
| evolutionaryscale/esm: THIRD_PARTY_NOTICE.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
Diagram steps
| evolutionaryscale/esm: README.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram title ESM3 workflow Individual claims | evolutionaryscale/esm: _assets/ESM3_README.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram title ESM3 workflow Individual claims | evolutionaryscale/esm: THIRD_PARTY_NOTICE.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Diagram title ESM3 workflow Individual claims | evolutionaryscale/esm: README.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
| Model type Multitrack generative protein transformer Individual claims | evolutionaryscale/esm: _assets/ESM3_README.md _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: bf343ba264b650dff7a073643725f9aaa1fdbe8d | source checked automated source review · 2026-09-16 Audit detailsInspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied. Field: Source artifact SHA-256: Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact |
Sources and history
Release 2026-10-22-1b7cf027e4f6 · Record review: discovered
5 source records and release history
- evolutionaryscale/esm official source · Original source · bf343ba264b650dff7a073643725f9aaa1fdbe8d
- evolutionaryscale/esm: README.md · Original source · bf343ba264b650dff7a073643725f9aaa1fdbe8d
- evolutionaryscale/esm: _assets/ESM3_README.md · Original source · bf343ba264b650dff7a073643725f9aaa1fdbe8d
- evolutionaryscale/esm: THIRD_PARTY_NOTICE.md · Original source · bf343ba264b650dff7a073643725f9aaa1fdbe8d
- evolutionaryscale/esm: LICENSE.md · Original source · bf343ba264b650dff7a073643725f9aaa1fdbe8d
Technical metadata and extraction receipts
Stable ID: discovery-model-esm3
- areas
- protein-structure
- access
- official_source_linked
- candidate benchmark ids
- discovery-benchmark-proteinbench
- entity level
- family
- reported name
- ESM3
- entity classification
- review date: 2026-09-17; rationale: The cited profile describes a named learned biological predictor or representation model/family. Preserve this identity separately from task-specific fitting, individual checkpoints, pipelines and hosted access.; source ids: evidence-official-4bf005200ebfb2586b78; evidence-official-0437be6150a3da9f8efc; evidence-official-1f027f8ecc53decdeec8; source locator: _assets/ESM3_README.md: overview, ESM3 Family, Running Locally and Licenses; ambiguities: None recorded
- missing metadata
- version: reason: unextracted
Related records
- applicable to: ProteinBench
- family: ESM3
- family: ESM3
- family: ESM3 (1.4B)
- uses model: MegSite + ESM3
- configuration of: ESM3 open (1.4B) [ProteinGym 144fe22b snapshot]