Model type
Protein representation transformer
Configuration
ESMC-300M supervised
ESM C learns protein sequence representations for downstream analysis.
Sources (5)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.mdHow it worksESMC workflow
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Sources (5)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.mdOverview
Inputs
Protein amino-acid sequences.
Outputs
Final-layer or all-layer protein representations and masked-token outputs.
Access
Official project documentation and implementation: https://github.com/evolutionaryscale/esm
Sources (5)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.mdlimited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Evaluations and results
16 evaluations · 32 results. Different protocols are not a single leaderboard.
Sorted by NDCG (cutoff not stated) (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.
All 32 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro to-P0A9X9; held-out test set · hydro Dataset subset: hydro | 0.078 ± 0.133 Spearman unitless · higher Uncertainty: ± 0.133 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro to-P0A9X9; held-out test set · hydro FLIP2 hydro to-P0A9X9; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A13; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 TrpB by-position; held-out test set · TrpB Dataset subset: TrpB | 0.980 ± 0.007 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.007 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 TrpB by-position; held-out test set · TrpB FLIP2 TrpB by-position; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A9; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 PDZ3 single-to-double; held-out test set · PDZ3 Dataset subset: PDZ3 | 0.499 ± 0.018 Spearman unitless · higher Uncertainty: ± 0.018 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 PDZ3 single-to-double; held-out test set · PDZ3 FLIP2 PDZ3 single-to-double; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A16; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 Amylase by-mutation; held-out test set · Amylase Dataset subset: Amylase | 0.061 ± 0.130 Spearman unitless · higher Uncertainty: ± 0.13 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 Amylase by-mutation; held-out test set · Amylase FLIP2 Amylase by-mutation; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A4; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro to-P01053; held-out test set · hydro Dataset subset: hydro | 0.949 ± 0.024 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.024 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro to-P01053; held-out test set · hydro FLIP2 hydro to-P01053; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A14; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 Amylase one-to-many; held-out test set · Amylase Dataset subset: Amylase | 0.869 ± 0.028 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.028 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 Amylase one-to-many; held-out test set · Amylase FLIP2 Amylase one-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A1; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro to-P06241; held-out test set · hydro Dataset subset: hydro | 0.934 ± 0.011 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.011 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro to-P06241; held-out test set · hydro FLIP2 hydro to-P06241; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A12; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro three-to-many; held-out test set · hydro Dataset subset: hydro | 0.977 ± 0.007 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.007 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro three-to-many; held-out test set · hydro FLIP2 hydro three-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A10; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 TrpB two-to-many; held-out test set · TrpB Dataset subset: TrpB | 0.991 ± 0.003 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.003 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 TrpB two-to-many; held-out test set · TrpB FLIP2 TrpB two-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A8; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 TrpB one-to-many; held-out test set · TrpB Dataset subset: TrpB | 0.306 ± 0.063 Spearman unitless · higher Uncertainty: ± 0.063 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 TrpB one-to-many; held-out test set · TrpB FLIP2 TrpB one-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A7; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 Amylase far-to-close; held-out test set · Amylase Dataset subset: Amylase | 0.011 ± 0.066 Spearman unitless · higher Uncertainty: ± 0.066 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 Amylase far-to-close; held-out test set · Amylase FLIP2 Amylase far-to-close; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A3; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 Amylase one-to-many; held-out test set · Amylase Dataset subset: Amylase | 0.148 ± 0.181 Spearman unitless · higher Uncertainty: ± 0.181 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 Amylase one-to-many; held-out test set · Amylase FLIP2 Amylase one-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A1; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 Amylase by-mutation; held-out test set · Amylase Dataset subset: Amylase | 0.793 ± 0.030 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.03 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 Amylase by-mutation; held-out test set · Amylase FLIP2 Amylase by-mutation; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A4; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 Amylase far-to-close; held-out test set · Amylase Dataset subset: Amylase | 0.819 ± 0.011 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.011 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 Amylase far-to-close; held-out test set · Amylase FLIP2 Amylase far-to-close; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A3; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro to-P06241; held-out test set · hydro Dataset subset: hydro | 0.161 ± 0.184 Spearman unitless · higher Uncertainty: ± 0.184 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro to-P06241; held-out test set · hydro FLIP2 hydro to-P06241; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A12; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro three-to-many; held-out test set · hydro Dataset subset: hydro | 0.399 ± 0.174 Spearman unitless · higher Uncertainty: ± 0.174 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro three-to-many; held-out test set · hydro FLIP2 hydro three-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A10; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro low-to-high; held-out test set · hydro Dataset subset: hydro | 0.099 ± 0.194 Spearman unitless · higher Uncertainty: ± 0.194 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro low-to-high; held-out test set · hydro FLIP2 hydro low-to-high; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A11; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro to-P0A9X9; held-out test set · hydro Dataset subset: hydro | 0.970 ± 0.001 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.001 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro to-P0A9X9; held-out test set · hydro FLIP2 hydro to-P0A9X9; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A13; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro to-P01053; held-out test set · hydro Dataset subset: hydro | 0.329 ± 0.190 Spearman unitless · higher Uncertainty: ± 0.19 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro to-P01053; held-out test set · hydro FLIP2 hydro to-P01053; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A14; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 IRED two-to-many; held-out test set · IRED Dataset subset: IRED | 0.954 ± 0.002 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.002 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 IRED two-to-many; held-out test set · IRED FLIP2 IRED two-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A5; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 PDZ3 single-to-double; held-out test set · PDZ3 Dataset subset: PDZ3 | 0.889 ± 0.011 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.011 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 PDZ3 single-to-double; held-out test set · PDZ3 FLIP2 PDZ3 single-to-double; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A16; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 TrpB one-to-many; held-out test set · TrpB Dataset subset: TrpB | 0.992 ± 0.001 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.001 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 TrpB one-to-many; held-out test set · TrpB FLIP2 TrpB one-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A7; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 NucB two-to-many; held-out test set · NucB Dataset subset: NucB | 0.723 ± 0.002 Spearman unitless · higher Uncertainty: ± 0.002 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 NucB two-to-many; held-out test set · NucB FLIP2 NucB two-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A6; row ESMC-300M supervised; column spearman |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 NucB two-to-many; held-out test set · NucB Dataset subset: NucB | 0.966 ± 0.000 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 NucB two-to-many; held-out test set · NucB FLIP2 NucB two-to-many; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A6; row ESMC-300M supervised; column ndcg |
| Configuration: ESMC-300M supervised | Protocol: FLIP2 hydro low-to-high; held-out test set · hydro Dataset subset: hydro | 0.870 ± 0.030 NDCG (cutoff not stated) unitless · higher Uncertainty: ± 0.03 (type not stated) Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceESMC-300M supervised · FLIP2 hydro low-to-high; held-out test set · hydro FLIP2 hydro low-to-high; held-out test set Aggregation: mean of five random-seed fits (Section 4.3) FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Table A11; row ESMC-300M supervised; column ndcg |
Source checking is not independent reproduction. Release 2026-11-02-a7f03c9e55a5.
Use this configuration
How it works, versions and access
Related profile: ESMC. This page retains the exact record and its evaluation context.
This configuration
ESMC-300M supervised: Family membership, checkpoint identity and aliases have not been inferred.
- record
- ESMC-300M supervised
- configuration
- Not reported
- entity type
- Configuration
How it works
How it works
Pre-normalized transformer with rotary embeddings, SwiGLU feed-forward activations and no linear/layer-norm biases. The documented inputs are protein amino-acid sequences. The output consists of final-layer or all-layer protein representations and masked-token outputs.
Sources (5)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.mdVersions and reproducibility
esmc-600m-2024-12 API identifier and biohub/ESMC-6B local example. The card specifies a2,048-token window after an initial 512-token training phase.
Sources (5)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.mdStrengths, limitations and unresolved questions
Strengths and limitations
Strengths and considerations
- The implementation supports local inference and a hosted interface, with downloadable model variants.
Sources (5)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md
Limitations and conditions
- Hosted API names and downloadable weights identify different versions. The main repository does not by itself provide a complete per-checkpoint training manifest.
Sources (5)
evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md
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-esmcSpecifications
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 | Protein representation transformerSources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Architecture | Pre-normalized transformer with rotary embeddings, SwiGLU feed-forward activations and no linear/layer-norm biases.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Inputs | Protein amino-acid sequences.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Outputs | Final-layer or all-layer protein representations and masked-token outputs.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Parameters | 300M/30 layers, 600M/36 layers and 6B/80 layers in the checked model card.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Known versions | esmc-600m-2024-12 API identifier and biohub/ESMC-6B local example.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Training data | UniRef, MGnify and JGI protein sequences clustered at 70% identity. The card distinguishes83M, 372M and 2B clusters respectively from the number of repeated training tokens.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Training cutoff | The official ESMC-6B card identifies UniRef, MGnify and JGI clusters and training stages, but does not give a latest-sequence date shared across those corpora. · Not reported in inspected sourcesSources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Context limits | The card specifies a2,048-token window after an initial 512-token training phase.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Weights licence | MIT is declared alongside third-party notices in the checked6B card; THIRD_PARTY_NOTICE.md lists dependency licences.Sources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Access | Official project documentation and implementation: https://github.com/evolutionaryscale/esmSources (5)evolutionaryscale/esm: README.md; evolutionaryscale/esm: _assets/ESM3_README.md; evolutionaryscale/esm: THIRD_PARTY_NOTICE.md; biohub/ESMC-6B: README.md; biohub/ESMC-6B: config.json · Official biohub/ESMC-6B card: Model Architecture, Parameters, Training Data, Training Procedure and Limitations; THIRD_PARTY_NOTICE.md |
| Code licence | MITSourcesevolutionaryscale/esm: LICENSE.md · LICENSE.md: licence text |
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.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: family discovery-model-esmc Individual claims | FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications Section 4.3 Fine-Tuned Protein Language Models; Figure 4 and ESMC-300M supervised tables Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256 | source checked ["automated-source-review"] · 2026-09-23 Audit detailsSource review establishes this relationship only. Exact evaluated configurations and original numerical review status remain unchanged. Pretrained ESMC-300M is fine-tuned for each FLIP2 split. This excludes the distinct naive/random-initialized control. Field: Claim: model-evaluation-identity-f1135e87eb0dcef979cf Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
Sources and history
Release 2026-11-02-a7f03c9e55a5 · Record review: discovered
1 source record and release history
- FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications · Original source · Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Technical metadata and extraction receipts
Stable ID: acquired-configuration-51e435f059eca0154ccf
- source locator
- Table A1; row ESMC-300M supervised; column spearman
- reported configuration
- ESMC-300M supervised
- missing metadata
- checkpoint revision: reason: unreported; model family: reason: unextracted
- provenance note
- Exact source-reported configuration or submission label. Family membership, checkpoint identity and aliases have not been inferred.
Related records
- family: ESMC
- system: ESMC-300M supervised · FLIP2 hydro to-P0A9X9; held-out test set · hydro
- system: ESMC-300M supervised · FLIP2 IRED two-to-many; held-out test set · IRED
- system: ESMC-300M supervised · FLIP2 rhomax by-wild-type; held-out test set · rhomax
- system: ESMC-300M supervised · FLIP2 hydro to-P06241; held-out test set · hydro
- system: ESMC-300M supervised · FLIP2 TrpB by-position; held-out test set · TrpB
- system: ESMC-300M supervised · FLIP2 NucB two-to-many; held-out test set · NucB
- system: ESMC-300M supervised · FLIP2 TrpB two-to-many; held-out test set · TrpB
- system: ESMC-300M supervised · FLIP2 TrpB one-to-many; held-out test set · TrpB
- system: ESMC-300M supervised · FLIP2 hydro three-to-many; held-out test set · hydro
- system: ESMC-300M supervised · FLIP2 PDZ3 single-to-double; held-out test set · PDZ3
- system: ESMC-300M supervised · FLIP2 hydro to-P01053; held-out test set · hydro
- system: ESMC-300M supervised · FLIP2 Amylase far-to-close; held-out test set · Amylase
- system: ESMC-300M supervised · FLIP2 Amylase close-to-far; held-out test set · Amylase
- system: ESMC-300M supervised · FLIP2 Amylase one-to-many; held-out test set · Amylase
- system: ESMC-300M supervised · FLIP2 Amylase by-mutation; held-out test set · Amylase
- system: ESMC-300M supervised · FLIP2 hydro low-to-high; held-out test set · hydro
- subject: ESMC-300M supervised: family ESMC