Configuration

ESM-1v (ensemble)

ESM-1v provides protein language models intended for zero-shot variant-effect scoring.

Sources (2)

facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models

1 evaluation · 5 results

How it worksESM-1v workflow

Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.

Sources (2)

facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models

Overview

Model type

Protein transformer for variant-effect scoring

Inputs

Protein amino-acid sequence and the specified sequence variation.

Outputs

Token probabilities used to calculate sequence-variant scores.

Access

Official project documentation and implementation: https://github.com/facebookresearch/esm

Sources (2)

facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models

limited source coverage · Automated source review, 2026-09-16. All specifications and missing details

Evaluations and results

1 evaluation · 5 results. Different protocols are not a single leaderboard.

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: ESM-1v (ensemble)	Task: ProteinGym ZS-SUB-AUC: Zero-shot substitutions, AUC Dataset subset: ProteinGym substitution DMS assays (ProteinGym split)	0.723 auc fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-1v (ensemble) on ProteinGym ZS-SUB-AUC: Zero-shot substitutions, AUC Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Aggregation: Not reported ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1v (ensemble)), column(Zero-shot substitutions, AUC)
Configuration: ESM-1v (ensemble)	Task: ProteinGym ZS-SUB-MCC: Zero-shot substitutions, MCC Dataset subset: ProteinGym substitution DMS assays (ProteinGym split)	0.32 mcc correlation · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-1v (ensemble) on ProteinGym ZS-SUB-MCC: Zero-shot substitutions, MCC Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Aggregation: Not reported ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1v (ensemble)), column(Zero-shot substitutions, MCC)
Configuration: ESM-1v (ensemble)	Task: ProteinGym ZS-SUB-NDCG: Zero-shot substitutions, NDCG@10% Dataset subset: ProteinGym substitution DMS assays (ProteinGym split)	0.749 ndcg fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-1v (ensemble) on ProteinGym ZS-SUB-NDCG: Zero-shot substitutions, NDCG@10% Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Aggregation: Not reported ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1v (ensemble)), column(Zero-shot substitutions, NDCG@10%)
Configuration: ESM-1v (ensemble)	Task: ProteinGym ZS-SUB-RECALL: Zero-shot substitutions, top 10% recall Dataset subset: ProteinGym substitution DMS assays (ProteinGym split)	0.211 recall fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-1v (ensemble) on ProteinGym ZS-SUB-RECALL: Zero-shot substitutions, top 10% recall Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Aggregation: Not reported ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1v (ensemble)), column(Zero-shot substitutions, top 10% recall)
Configuration: ESM-1v (ensemble)	Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Dataset subset: ProteinGym substitution DMS assays (ProteinGym split)	0.407 spearman correlation · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source ESM-1v (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Aggregation: Not reported ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1v (ensemble)), column(Zero-shot substitutions, Spearman)

Source checking is not independent reproduction. Release 2026-09-29-06401fd5b220.

Use this model

How it works, versions and access

Related profile: ESM-1v. This page retains the exact record and its evaluation context.

This configuration

Fitness prediction model evaluated by the ProteinGym authors under their harness.

record: ESM-1v (ensemble)
configuration: Not reported
entity type: Configuration

How it works

ESM-1v provides protein language models intended for zero-shot variant-effect scoring. Transformer architecture shared with ESM-1b, trained on UniRef90; multiple released model instances support the variant-scoring workflow. The documented inputs are protein amino-acid sequence and the specified sequence variation. The output consists of token probabilities used to calculate sequence-variant scores.

Sources (2)

facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models

Versions and reproducibility

esm1v_t33_650M_UR90S_1 through _5. The applicable input limits require configuration-specific checking.

Sources (2)

facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths and considerations

The package provides a dedicated variant-prediction example without fitting a task-specific supervised classifier.
Sources (2)
facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models

Limitations and conditions

The five released instances are distinct models. A language-model score is not a calibrated clinical pathogenicity probability.
Sources (2)
facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models

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-esm-1v

Specifications

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.

Inputs, outputs and configuration
Property	Description and evidence
Model type	Protein transformer for variant-effect scoring Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Architecture	Transformer architecture shared with ESM-1b, trained on UniRef90; multiple released model instances support the variant-scoring workflow. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Inputs	Protein amino-acid sequence and the specified sequence variation. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Outputs	Token probabilities used to calculate sequence-variant scores. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Parameters	650 million per model; 33 transformer layers. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Known versions	esm1v_t33_650M_UR90S_1 through _5. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Training data	UniRef90/S 2020_03, as reported in the pretrained-model table. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Training cutoff	The official model table identifies the UniRef90/S2020_03 training release; this is a corpus version rather than a verified latest-sequence deposition date. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Context limits	The inspected official esm1v_t33_650M_UR90S_1 config reserves 1,026 absolute positions. This includes model positions and is not a claim of 1,026-residue validated biological context. Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Weights licence	Separate checkpoint-distribution terms are not stated in the inspected release documentation and licence material. The source-code licence alone is not recorded as an explicit weight grant. · Not reported in inspected sources Sources (3) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json; facebookresearch/esm: LICENSE · README.md: Main models, Zero-shot variant prediction and Pre-trained Models; LICENSE: licence text
Access	Official project documentation and implementation: https://github.com/facebookresearch/esm Sources (2) facebookresearch/esm: README.md; facebook/esm1v_t33_650M_UR90S_1: config.json · README.md: Main models, Zero-shot variant prediction and Pre-trained Models
Code licence	MIT Sources facebookresearch/esm: LICENSE · LICENSE: 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

Claims, original sources and review scope · Release 2026-09-29-06401fd5b220
Property and statement	Original source and location	Review and provenance
Relationship: family discovery-model-esm-1v Individual claims	ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design Original source ↗ Section 4.1 Zero-shot benchmarks; Table 2 named model row Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256 Retrieved: 2026-09-17T08:06:28.522924+00:00	source checked automated source review · 2026-09-23 Audit details Source review establishes this relationship only. Exact evaluated configurations and original numerical review status remain unchanged. The primary table and baseline descriptions identify the evaluated family. ESM-1v remains an ensemble; ESM-2 remains 15B; inverse-folding conditions remain distinct. Field: `links:family:discovery-model-esm-1v` Claim: model-evaluation-identity-b6431ad18a12892be1fa Source artifact SHA-256: `a3b08cc4a6befd64620cf0f287d78d55a36dc2639955f52c5655f83833c50104` Hash scope: Exact retrieved primary paper artifact bytes. Inspected artifact

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

Release 2026-09-29-06401fd5b220 · Record review: source checked

1 source records and release history

ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Original source · Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256

Download this release

Technical metadata and extraction receipts

Stable ID: proteingym-method-esm-1v-ensemble

areas: proteins-complexes
source locator: Table 2, row(ESM-1v (ensemble))
missing metadata: checkpoint revision: unreported; parameters: unextracted

Related records

Suggest a correction