rewirebio.iobenchmarks
Configuration

ESM-2 3B (HyPLM report)

Runs the public ESM-2 3B checkpoint in BF16 through Hugging Face Transformers to produce embeddings, probe features and masked-language-model scores.

6 evaluations · 6 results

Overview

An explanatory profile has not yet been reviewed for this record.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

Model Capabilities & Hardware Requirements

Deployment facts recorded for ESM-2 3B (HyPLM report). Fields the record does not carry are marked not recorded.

ParametersNot recorded
Minimum VRAMNot recorded
Inference LatencyNot recorded
Context WindowNot recorded
AccessNot recorded
LicenseNot recorded

Evaluated Checkpoint: Public ESM-2 3B checkpoint; no revision printed

Evaluations and results

6 evaluations · 6 results. Different protocols are not a single leaderboard.

Sorted by Log loss (Mean negative log-likelihood of masked residues in nats on the report's held-out UniRef panel, identical masks and crops for every model) (lower 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 configurationProtocol and datasetLog loss (Mean negative log-likelihood of masked residues in nats on the report's held-out UniRef panel, identical masks and crops for every model)Log loss (Mean negative log-likelihood of masked residues in nats on the report's recent-PDB panel, identical masks and crops for every model)Perplexity (Masked-residue perplexity on the held-out UniRef panel: the exponential of the mean negative log-likelihood of masked residues under the model itself, with identical masks and crops for every model)Perplexity (Masked-residue perplexity on the recent PDB chains: the exponential of the mean negative log-likelihood of masked residues under the model itself, with identical masks and crops for every model)Spearman (Spearman correlation averaged over the 217 substitution assays of ProteinGym v1.3 with the benchmark's official aggregation, from masked-marginal zero-shot scores; per-assay values rounded to three decimals before aggregation)Spearman (Unweighted mean of the test Spearman correlations on the ten regression splits Appendix A uses for summary statistics, excluding GB1 two-vs-rest, GB1 three-vs-rest and the FLIP2 hydrolase split)
ESM-2 3B (HyPLM report)EvaluationNLL, held-out UniRef panel (HyPLM report), Held-out UniRef panel (479 proteins, HyPLM report)1.67Not reportedNot reportedNot reportedNot reportedNot reported
ESM-2 3B (HyPLM report)EvaluationNLL, recent PDB chains (HyPLM report), Recent PDB chains (887, HyPLM report)Not reported1.33Not reportedNot reportedNot reportedNot reported
ESM-2 3B (HyPLM report)EvaluationMasked-residue perplexity, held-out UniRef panel (HyPLM report), Held-out UniRef panel (479 proteins, HyPLM report)Not reportedNot reported5.3Not reportedNot reportedNot reported
ESM-2 3B (HyPLM report)EvaluationMasked-residue perplexity, recent PDB chains (HyPLM report), Recent PDB chains (887, HyPLM report)Not reportedNot reportedNot reported3.77Not reportedNot reported
ESM-2 3B (HyPLM report)EvaluationProteinGym zero-shot (HyPLM report), ProteinGym DMS substitution assays at commit 144fe22 (217 assays)Not reportedNot reportedNot reportedNot reported0.407Not reported
ESM-2 3B (HyPLM report)EvaluationMean Spearman over ten regression splits (HyPLM report), Ten regression splits for summary statistics (HyPLM report)Not reportedNot reportedNot reportedNot reportedNot reported0.573
All 6 result rows with coverage, uncertainty and sources
Filter evaluations

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configurationProtocol and datasetFindingEvidence and details
Configuration: ESM-2 3B (HyPLM report)Protocol: NLL, held-out UniRef panel (HyPLM report)
Dataset: Held-out UniRef panel (479 proteins, HyPLM report)
1.67 Log loss (Mean negative log-likelihood of masked residues in nats on the report's held-out UniRef panel, identical masks and crops for every model)
unitless · lower

Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

ESM-2 3B on nll-held-out-uniref (HyPLM report)

hyplm-flash-20261106-protocol-nll-held-out-uniref

Aggregation: Not reported

HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 3, row 'ESM-2 3B', column 'Held-out UniRef proteins, NLL' (page 6 of 13)
Configuration: ESM-2 3B (HyPLM report)Protocol: NLL, recent PDB chains (HyPLM report)
Dataset: Recent PDB chains (887, HyPLM report)
1.33 Log loss (Mean negative log-likelihood of masked residues in nats on the report's recent-PDB panel, identical masks and crops for every model)
unitless · lower

Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

ESM-2 3B on nll-recent-pdb (HyPLM report)

hyplm-flash-20261106-protocol-nll-recent-pdb

Aggregation: Not reported

HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 3, row 'ESM-2 3B', column 'Recent PDB chains, NLL' (page 6 of 13)
Configuration: ESM-2 3B (HyPLM report)Protocol: Masked-residue perplexity, held-out UniRef panel (HyPLM report)
Dataset: Held-out UniRef panel (479 proteins, HyPLM report)
5.3 Perplexity (Masked-residue perplexity on the held-out UniRef panel: the exponential of the mean negative log-likelihood of masked residues under the model itself, with identical masks and crops for every model)
unitless · lower

Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

ESM-2 3B on ppl-held-out-uniref (HyPLM report)

hyplm-flash-20261106-protocol-ppl-held-out-uniref

Aggregation: Not reported

HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 3, row 'ESM-2 3B', column 'Held-out UniRef proteins, PPL' (page 6 of 13)
Configuration: ESM-2 3B (HyPLM report)Protocol: Masked-residue perplexity, recent PDB chains (HyPLM report)
Dataset: Recent PDB chains (887, HyPLM report)
3.77 Perplexity (Masked-residue perplexity on the recent PDB chains: the exponential of the mean negative log-likelihood of masked residues under the model itself, with identical masks and crops for every model)
unitless · lower

Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

ESM-2 3B on ppl-recent-pdb (HyPLM report)

hyplm-flash-20261106-protocol-ppl-recent-pdb

Aggregation: Not reported

HyPLM Technical Report (Hyra Team, 2026-10-12) · Table 3, row 'ESM-2 3B', column 'Recent PDB chains, PPL' (page 6 of 13)
Configuration: ESM-2 3B (HyPLM report)Protocol: ProteinGym zero-shot (HyPLM report)
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.407 Spearman (Spearman correlation averaged over the 217 substitution assays of ProteinGym v1.3 with the benchmark's official aggregation, from masked-marginal zero-shot scores; per-assay values rounded to three decimals before aggregation)
unitless · higher

Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

ESM-2 3B on proteingym-zero-shot (HyPLM report)

hyplm-flash-20261106-protocol-proteingym-zero-shot

Aggregation: Not reported

HyPLM Technical Report (Hyra Team, 2026-10-12) · Figure 3a, bar 'ESM-2 3B', printed value label (page 5 of 13)
Configuration: ESM-2 3B (HyPLM report)Protocol: Mean Spearman over ten regression splits (HyPLM report)
Dataset: Ten regression splits for summary statistics (HyPLM report)
0.573 Spearman (Unweighted mean of the test Spearman correlations on the ten regression splits Appendix A uses for summary statistics, excluding GB1 two-vs-rest, GB1 three-vs-rest and the FLIP2 hydrolase split)
unitless · higher

Uncertainty: Not reported by the source: Section 4 says intervals are paired bootstrap intervals; none is printed here.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

ESM-2 3B on regression-ten-split-average (HyPLM report)

hyplm-flash-20261106-protocol-regression-ten-split-average

Aggregation: Not reported

HyPLM Technical Report (Hyra Team, 2026-10-12) · Section 4.2, sentence 'Averaged over the ten regression splits ...', value for ESM-2 3B (page 4 of 13)

Source checking is not independent reproduction. Release 2026-11-21-c8bea25e5abb.

Use this configuration

How it works, versions and access

Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.

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.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-c8bea25e5abb
Property and statementOriginal source and locationReview and provenance

No evidence rows match these filters. Choose another scope or clear the search.

Sources and history

Release 2026-11-21-c8bea25e5abb · Record review: source checked

1 source record and release historyDownload this release (gzip)
Technical metadata and record identifiers

Technical metadata for this record is tracked under stable ID hyplm-flash-20261106-config-esm-2-3b.

Source locator: Appendix A 'Models' and 'Deviations'; Section 4

Related records

How to cite

ESM-2 3B (HyPLM report). Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/hyplm-flash-20261106-config-esm-2-3b/

Release 2026-11-21-c8bea25e5abb. Rankings and values change between releases, so keep the release id in the citation. Cite the original publication as well for any value you reuse: each result links its source.

The compilation and annotations made by Rewire in this release are offered under CC BY 4.0; values and text taken from cited sources stay under the terms of those sources and are not covered by this licence. Data licence

Suggest a correction