rewirebio.iobenchmarks
Configuration

RFD4-Proteina (high thinking), previous release pair

Configures RFD4-Proteina for one-shot generation with the high-thinking recycling schedule on the flow and autoencoder pair that preceded the 2026-10-19 release.

4 evaluations · 4 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 RFD4-Proteina (high thinking), previous release pair. 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: The release pair preceding 2026-10-19, as the model card names it; no checkpoint filename is tied to the efficiency table

Evaluations and results

4 evaluations · 4 results. Different protocols are not a single leaderboard.

Sorted by Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: AF3 binder pTM at least 80 on the native 0 to 100 scale, minimum interface PAE at most 1.5 A, and target-backbone-aligned binder backbone RMSD at most 2.5 A, with valid required target mapping and conditioning) (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 datasetPassing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: AF3 binder pTM at least 80 on the native 0 to 100 scale, minimum interface PAE at most 1.5 A, and target-backbone-aligned binder backbone RMSD at most 2.5 A, with valid required target mapping and conditioning)Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: AF3 minimum interface PAE at most 1.5 A and DNA-target-aligned binder backbone RMSD at most 8 A, aligning on all target DNA atoms in both the fixed and unfixed modes)Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: AF3 minimum interface PAE at most 1.5 A, chain-of-interest ipTM at least 80 out of 100, protein-backbone RMSD at most 1.5 A, and protein-backbone-aligned ligand RMSD at most 5 A (the sm_binder criterion, not the looser historical disco criterion))Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: protein-backbone RMSD at most 2 A, required motif all-atom RMSD at most 1.5 A, and AF3 minimum interface PAE below 1.5 A)
RFD4-Proteina (high thinking), previous release pairEvaluationFrozen project benchmark protein binding, backbone conditioning: distinct passers per GPU-hour, Frozen project benchmark protein-binding target set, 24 targets71.6Not reportedNot reportedNot reported
RFD4-Proteina (high thinking), previous release pairEvaluationFrozen project benchmark Atomic Motif Enzyme, fixed ligand: distinct passers per GPU-hour, Frozen project benchmark Atomic Motif Enzyme target set, 37 tasksNot reportedNot reportedNot reported91.2
RFD4-Proteina (high thinking), previous release pairEvaluationFrozen project benchmark DNA binding, unfixed DNA: distinct passers per GPU-hour, Frozen project benchmark DNA-binding target set, 3 targetsNot reported42Not reportedNot reported
RFD4-Proteina (high thinking), previous release pairEvaluationFrozen project benchmark small-molecule binding, unfixed ligand: distinct passers per GPU-hour, Frozen project benchmark small-molecule binding target set, 77 targetsNot reportedNot reported112Not reported
All 4 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: RFD4-Proteina (high thinking), previous release pairProtocol: Frozen project benchmark Atomic Motif Enzyme, fixed ligand: distinct passers per GPU-hour
Dataset: Frozen project benchmark Atomic Motif Enzyme target set, 37 tasks
91.2 Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: protein-backbone RMSD at most 2 A, required motif all-atom RMSD at most 1.5 A, and AF3 minimum interface PAE below 1.5 A)
count-per-hour · higher

Uncertainty: Not reported by the source: The table prints point values with no interval or spread.

Coverage: 7400/7400 design attempts scored (37 targets times 200 original attempts)

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

RFD4-Proteina, high thinking on Frozen project benchmark Atomic Motif Enzyme, fixed ligand: distinct passers per GPU-hour

rfd4-proteina-20261115-protocol-ame-fixed-ligand-passers-per-gpu-hour

Aggregation: Arithmetic mean of target-level rates

RFD4-Proteina Hugging Face model card at revision 621b785; RFD4-Proteina benchmarks/summary.csv at revision 621b785 · Hugging Face model card, 'Measured design efficiency' table, row 'Atomic Motif Enzyme', column 'RFD4-Proteina, high thinking'
Configuration: RFD4-Proteina (high thinking), previous release pairProtocol: Frozen project benchmark DNA binding, unfixed DNA: distinct passers per GPU-hour
Dataset: Frozen project benchmark DNA-binding target set, 3 targets
42 Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: AF3 minimum interface PAE at most 1.5 A and DNA-target-aligned binder backbone RMSD at most 8 A, aligning on all target DNA atoms in both the fixed and unfixed modes)
count-per-hour · higher

Uncertainty: Not reported by the source: The table prints point values with no interval or spread.

Coverage: 600/600 design attempts scored (3 targets times 200 original attempts)

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

RFD4-Proteina, high thinking on Frozen project benchmark DNA binding, unfixed DNA: distinct passers per GPU-hour

rfd4-proteina-20261115-protocol-dna-unfixed-passers-per-gpu-hour

Aggregation: Arithmetic mean of target-level rates

RFD4-Proteina Hugging Face model card at revision 621b785; RFD4-Proteina benchmarks/summary.csv at revision 621b785 · Hugging Face model card, 'Measured design efficiency' table, row 'DNA binders', column 'RFD4-Proteina, high thinking'
Configuration: RFD4-Proteina (high thinking), previous release pairProtocol: Frozen project benchmark protein binding, backbone conditioning: distinct passers per GPU-hour
Dataset: Frozen project benchmark protein-binding target set, 24 targets
71.6 Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: AF3 binder pTM at least 80 on the native 0 to 100 scale, minimum interface PAE at most 1.5 A, and target-backbone-aligned binder backbone RMSD at most 2.5 A, with valid required target mapping and conditioning)
count-per-hour · higher

Uncertainty: Not reported by the source: The table prints point values with no interval or spread.

Coverage: 4800/4800 design attempts scored (24 targets times 200 original attempts)

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

RFD4-Proteina, high thinking on Frozen project benchmark protein binding, backbone conditioning: distinct passers per GPU-hour

rfd4-proteina-20261115-protocol-ppi-backbone-passers-per-gpu-hour

Aggregation: Arithmetic mean of target-level rates

RFD4-Proteina Hugging Face model card at revision 621b785; RFD4-Proteina benchmarks/summary.csv at revision 621b785 · Hugging Face model card, 'Measured design efficiency' table, row 'Protein binders', column 'RFD4-Proteina, high thinking'
Configuration: RFD4-Proteina (high thinking), previous release pairProtocol: Frozen project benchmark small-molecule binding, unfixed ligand: distinct passers per GPU-hour
Dataset: Frozen project benchmark small-molecule binding target set, 77 targets
112 Passing design clusters per accelerator hour (mean over targets of structurally distinct computational passers per GPU-hour, counting generation and common AlphaFold3 evaluation time, with 200 original attempts per target and failed attempts included; passing designs clustered within each model, protocol and target by Foldseek at TM 0.5; pass criterion: AF3 minimum interface PAE at most 1.5 A, chain-of-interest ipTM at least 80 out of 100, protein-backbone RMSD at most 1.5 A, and protein-backbone-aligned ligand RMSD at most 5 A (the sm_binder criterion, not the looser historical disco criterion))
count-per-hour · higher

Uncertainty: Not reported by the source: The table prints point values with no interval or spread.

Coverage: 15400/15400 design attempts scored (77 targets times 200 original attempts)

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

RFD4-Proteina, high thinking on Frozen project benchmark small-molecule binding, unfixed ligand: distinct passers per GPU-hour

rfd4-proteina-20261115-protocol-small-molecule-unfixed-passers-per-gpu-hour

Aggregation: Arithmetic mean of target-level rates

RFD4-Proteina Hugging Face model card at revision 621b785; RFD4-Proteina benchmarks/summary.csv at revision 621b785 · Hugging Face model card, 'Measured design efficiency' table, row 'Small-molecule binders', column 'RFD4-Proteina, high thinking'

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

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

Technical metadata for this record is tracked under stable ID rfd4-proteina-20261115-config-previous-release-high-thinking.

Source locator: Hugging Face model card, 'Measured design efficiency' scope note and table, 'Checkpoints in this release' table and 'Compute presets' table; benchmarks/METHODS.md, 'Model recipes and adaptations' row RFD4-Proteina

Related records

How to cite

RFD4-Proteina (high thinking), previous release pair. Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/rfd4-proteina-20261115-config-previous-release-high-thinking/

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