rewirebio.iobenchmarks
Configuration

Caduceus, largest reverse-complement-equivariant model, frozen embeddings

The largest reverse-complement-equivariant Caduceus model, used as a frozen embedding extractor for the BEND tasks.

7 evaluations · 7 results5 open disagreements between studies involve its model

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 Caduceus, largest reverse-complement-equivariant model, frozen embeddings. 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

Evaluations and results

7 evaluations · 7 results. Different protocols are not a single leaderboard.

Sorted by AUPRC (BEND Enhancer Annotation, two-layer CNN trained on frozen embeddings) (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 configurationProtocol and datasetAUPRC (BEND Enhancer Annotation, two-layer CNN trained on frozen embeddings)AUROC (BEND Chromatin Accessibility, two-layer CNN trained on frozen embeddings)AUROC (BEND CpG Methylation, two-layer CNN trained on frozen embeddings)AUROC (BEND Disease VEP, cosine distance between the embeddings of the reference and alternative sequences)AUROC (BEND Expression VEP, cosine distance between the embeddings of the reference and alternative sequences)AUROC (BEND Histone Modification, two-layer CNN trained on frozen embeddings)MCC (BEND Gene Finding, two-layer CNN trained on frozen embeddings)
Caduceus, largest reverse-complement-equivariant model, frozen embeddingsEvaluationBEND ENHANCER: Enhancer annotation, Fulco 2019, Gasperini 2019 and Enformer enhancer set (BEND split)0.03Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
Caduceus, largest reverse-complement-equivariant model, frozen embeddingsEvaluationBEND CHROMATIN: Chromatin accessibility, ENCODE chromatin accessibility (BEND split)Not reported0.79Not reportedNot reportedNot reportedNot reportedNot reported
Caduceus, largest reverse-complement-equivariant model, frozen embeddingsEvaluationBEND CPG: CpG methylation, ENCODE CpG methylation (BEND split)Not reportedNot reported0.91Not reportedNot reportedNot reportedNot reported
Caduceus, largest reverse-complement-equivariant model, frozen embeddingsEvaluationBEND VARIANT-DISEASE: Noncoding variant effects on disease, ClinVar disease variants (BEND split)Not reportedNot reportedNot reported0.54Not reportedNot reportedNot reported
Caduceus, largest reverse-complement-equivariant model, frozen embeddingsEvaluationBEND VARIANT-EXPRESSION: Noncoding variant effects on expression, DeepSEA expression variants (BEND split)Not reportedNot reportedNot reportedNot reported0.53Not reportedNot reported
Caduceus, largest reverse-complement-equivariant model, frozen embeddingsEvaluationBEND GENE-FINDING: Gene finding, GENCODE (BEND split)Not reportedNot reportedNot reportedNot reportedNot reportedNot reported0.54
Caduceus, largest reverse-complement-equivariant model, frozen embeddingsEvaluationBEND HISTONE: Histone modification, ENCODE histone modification (BEND split)Not reportedNot reportedNot reportedNot reportedNot reported0.78Not reported
All 7 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: Caduceus, largest reverse-complement-equivariant model, frozen embeddingsTask: BEND CHROMATIN: Chromatin accessibility
Dataset subset: ENCODE chromatin accessibility (BEND split)
0.79 AUROC (BEND Chromatin Accessibility, two-layer CNN trained on frozen embeddings)
unitless · higher

Uncertainty: Not reported by the source: Table 2 prints no interval or spread.

Coverage: Not reported scored / Not reported eligible

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

Caduceus on BEND Chromatin Accessibility

bend-task-chromatin

Aggregation: Not reported

A Phylogenetic Approach to Genomic Language Modeling · Table 2, row 'Caduceus', column 'Chromatin Accessibility (AUROC)'
Configuration: Caduceus, largest reverse-complement-equivariant model, frozen embeddingsTask: BEND CPG: CpG methylation
Dataset subset: ENCODE CpG methylation (BEND split)
0.91 AUROC (BEND CpG Methylation, two-layer CNN trained on frozen embeddings)
unitless · higher

Uncertainty: Not reported by the source: Table 2 prints no interval or spread.

Coverage: Not reported scored / Not reported eligible

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

Caduceus on BEND CpG Methylation

bend-task-cpg

Aggregation: Not reported

A Phylogenetic Approach to Genomic Language Modeling · Table 2, row 'Caduceus', column 'CpG Methylation (AUROC)'
Configuration: Caduceus, largest reverse-complement-equivariant model, frozen embeddingsTask: BEND VARIANT-DISEASE: Noncoding variant effects on disease
Dataset subset: ClinVar disease variants (BEND split)
0.54 AUROC (BEND Disease VEP, cosine distance between the embeddings of the reference and alternative sequences)
unitless · higher

Uncertainty: Not reported by the source: Table 2 prints no interval or spread.

Coverage: Not reported scored / Not reported eligible

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

Caduceus on BEND Disease VEP

bend-task-variant-disease

Aggregation: Not reported

A Phylogenetic Approach to Genomic Language Modeling · Table 2, row 'Caduceus', column 'Disease VEP (AUROC)'
Configuration: Caduceus, largest reverse-complement-equivariant model, frozen embeddingsTask: BEND ENHANCER: Enhancer annotation
Dataset subset: Fulco 2019, Gasperini 2019 and Enformer enhancer set (BEND split)
0.03 AUPRC (BEND Enhancer Annotation, two-layer CNN trained on frozen embeddings)
unitless · higher

Uncertainty: Not reported by the source: Table 2 prints no interval or spread.

Coverage: Not reported scored / Not reported eligible

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

Caduceus on BEND Enhancer Annotation

bend-task-enhancer

Aggregation: Not reported

A Phylogenetic Approach to Genomic Language Modeling · Table 2, row 'Caduceus', column 'Enhancer Annotation (AUPRC)'
Configuration: Caduceus, largest reverse-complement-equivariant model, frozen embeddingsTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
Dataset subset: DeepSEA expression variants (BEND split)
0.53 AUROC (BEND Expression VEP, cosine distance between the embeddings of the reference and alternative sequences)
unitless · higher

Uncertainty: Not reported by the source: Table 2 prints no interval or spread.

Coverage: Not reported scored / Not reported eligible

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

Caduceus on BEND Expression VEP

bend-task-variant-expression

Aggregation: Not reported

A Phylogenetic Approach to Genomic Language Modeling · Table 2, row 'Caduceus', column 'Expression VEP (AUROC)'
Configuration: Caduceus, largest reverse-complement-equivariant model, frozen embeddingsTask: BEND GENE-FINDING: Gene finding
Dataset subset: GENCODE (BEND split)
0.54 MCC (BEND Gene Finding, two-layer CNN trained on frozen embeddings)
unitless · higher

Uncertainty: Not reported by the source: Table 2 prints no interval or spread.

Coverage: Not reported scored / Not reported eligible

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

Caduceus on BEND Gene Finding

bend-task-gene-finding

Aggregation: Not reported

A Phylogenetic Approach to Genomic Language Modeling · Table 2, row 'Caduceus', column 'Gene Finding (MCC)'
Configuration: Caduceus, largest reverse-complement-equivariant model, frozen embeddingsTask: BEND HISTONE: Histone modification
Dataset subset: ENCODE histone modification (BEND split)
0.78 AUROC (BEND Histone Modification, two-layer CNN trained on frozen embeddings)
unitless · higher

Uncertainty: Not reported by the source: Table 2 prints no interval or spread.

Coverage: Not reported scored / Not reported eligible

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

Caduceus on BEND Histone Modification

bend-task-histone

Aggregation: Not reported

A Phylogenetic Approach to Genomic Language Modeling · Table 2, row 'Caduceus', column 'Histone Modification (AUROC)'

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 phylogpn-20261116-config-caduceus-largest-rce.

Source locator: Results, Embedding Evaluation (paragraph 2); Table 2 row 'Caduceus'

Related records

How to cite

Caduceus, largest reverse-complement-equivariant model, frozen embeddings. Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/phylogpn-20261116-config-caduceus-largest-rce/

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