Co-generated structures and sequences at fixed lengths (ProteinBench split)

Subset and evaluation context

This record describes a particular subset or cohort used in an evaluation. Its results do not describe the full dataset.

Evaluation results

Release 2026-09-17-134cd1815de8 · 76 evaluations · 76 metric rows. Different protocols are not a single leaderboard. Where several source tables report the same metric, the published comparisons above offer a pooled view that names what it does not hold constant.

Evaluation setupMetricEvidence origin

Results grouped by the exact reported evaluation
Metric and finding	Coverage and uncertainty	Evidence
ESM3 on ProteinBench CO-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust. Configuration: ESM3Task: ProteinBench CO-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust.Dataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.64 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 100, Diversity Max Clust. ↑) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TM Configuration: ESM3Task: ProteinBench CO-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.41 max-tm Unit: score · Direction: lower	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 100, Novelty Max TM ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-100-QUALITY-SCRMSD: length 100, Quality scRMSD Configuration: ESM3Task: ProteinBench CO-LENGTH-100-QUALITY-SCRMSD: length 100, Quality scRMSDDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
13.80 ± 10.51 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 10.51 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 100, Quality scRMSD ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-100-QUALITY-SCTM: length 100, Quality scTM Configuration: ESM3Task: ProteinBench CO-LENGTH-100-QUALITY-SCTM: length 100, Quality scTMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.72 ± 0.19 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.19 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 100, Quality scTM ↑) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust. Configuration: ESM3Task: ProteinBench CO-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust.Dataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.63 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 200, Diversity Max Clust. ↑) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TM Configuration: ESM3Task: ProteinBench CO-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.61 max-tm Unit: score · Direction: lower	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 200, Novelty Max TM ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-200-QUALITY-SCRMSD: length 200, Quality scRMSD Configuration: ESM3Task: ProteinBench CO-LENGTH-200-QUALITY-SCRMSD: length 200, Quality scRMSDDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
21.18 ± 16.19 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 16.19 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 200, Quality scRMSD ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-200-QUALITY-SCTM: length 200, Quality scTM Configuration: ESM3Task: ProteinBench CO-LENGTH-200-QUALITY-SCTM: length 200, Quality scTMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.63 ± 0.20 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.20 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 200, Quality scTM ↑) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust. Configuration: ESM3Task: ProteinBench CO-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust.Dataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.52 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 300, Diversity Max Clust. ↑) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TM Configuration: ESM3Task: ProteinBench CO-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.73 max-tm Unit: score · Direction: lower	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 300, Novelty Max TM ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-300-QUALITY-SCRMSD: length 300, Quality scRMSD Configuration: ESM3Task: ProteinBench CO-LENGTH-300-QUALITY-SCRMSD: length 300, Quality scRMSDDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
25.5 ± 20.68 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 20.68 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 300, Quality scRMSD ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-300-QUALITY-SCTM: length 300, Quality scTM Configuration: ESM3Task: ProteinBench CO-LENGTH-300-QUALITY-SCTM: length 300, Quality scTMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.59 ± 0.21 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.21 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 300, Quality scTM ↑) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-500-DIVERSITY-MAX-CLUST: length 500, Diversity Max Clust. Configuration: ESM3Task: ProteinBench CO-LENGTH-500-DIVERSITY-MAX-CLUST: length 500, Diversity Max Clust.Dataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.46 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 500, Diversity Max Clust. ↑) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-500-NOVELTY-MAX-TM: length 500, Novelty Max TM Configuration: ESM3Task: ProteinBench CO-LENGTH-500-NOVELTY-MAX-TM: length 500, Novelty Max TMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.78 max-tm Unit: score · Direction: lower	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 500, Novelty Max TM ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-500-QUALITY-SCRMSD: length 500, Quality scRMSD Configuration: ESM3Task: ProteinBench CO-LENGTH-500-QUALITY-SCRMSD: length 500, Quality scRMSDDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
26.72 ± 21.08 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 21.08 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 500, Quality scRMSD ↓) Source checking is not independent reproduction.
ESM3 on ProteinBench CO-LENGTH-500-QUALITY-SCTM: length 500, Quality scTM Configuration: ESM3Task: ProteinBench CO-LENGTH-500-QUALITY-SCTM: length 500, Quality scTMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.64 ± 0.20 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.20 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(ESM3*), column(length 500, Quality scTM ↑) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust. Configuration: MultiflowTask: ProteinBench CO-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust.Dataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.33 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 100, Diversity Max Clust. ↑) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TM Configuration: MultiflowTask: ProteinBench CO-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.71 max-tm Unit: score · Direction: lower	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 100, Novelty Max TM ↓) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-100-QUALITY-SCRMSD: length 100, Quality scRMSD Configuration: MultiflowTask: ProteinBench CO-LENGTH-100-QUALITY-SCRMSD: length 100, Quality scRMSDDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
1.10 ± 0.71 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.71 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 100, Quality scRMSD ↓) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-100-QUALITY-SCTM: length 100, Quality scTM Configuration: MultiflowTask: ProteinBench CO-LENGTH-100-QUALITY-SCTM: length 100, Quality scTMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.96 ± 0.04 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.04 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 100, Quality scTM ↑) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust. Configuration: MultiflowTask: ProteinBench CO-LENGTH-200-DIVERSITY-MAX-CLUST: length 200, Diversity Max Clust.Dataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.42 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 200, Diversity Max Clust. ↑) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TM Configuration: MultiflowTask: ProteinBench CO-LENGTH-200-NOVELTY-MAX-TM: length 200, Novelty Max TMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.71 max-tm Unit: score · Direction: lower	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 200, Novelty Max TM ↓) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-200-QUALITY-SCRMSD: length 200, Quality scRMSD Configuration: MultiflowTask: ProteinBench CO-LENGTH-200-QUALITY-SCRMSD: length 200, Quality scRMSDDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
1.61 ± 1.73 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 1.73 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 200, Quality scRMSD ↓) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-200-QUALITY-SCTM: length 200, Quality scTM Configuration: MultiflowTask: ProteinBench CO-LENGTH-200-QUALITY-SCTM: length 200, Quality scTMDataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.95 ± 0.04 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.04 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 200, Quality scTM ↑) Source checking is not independent reproduction.
Multiflow on ProteinBench CO-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust. Configuration: MultiflowTask: ProteinBench CO-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust.Dataset subset: Co-generated structures and sequences at fixed lengths (ProteinBench split) Joint structure and sequence generation at a fixed length. Values are the mean and standard deviation over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.58 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 5, row(Multiflow), column(length 300, Diversity Max Clust. ↑) Source checking is not 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.

2 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-17-134cd1815de8
Property and statement	Original source and location	Review and provenance
description The split of Co-generated structures and sequences at fixed lengths that ProteinBench evaluated on. The upstream dataset release is not catalogued here, so no claim is made that this matches its original splits. Context-only references	proteinbench primary benchmark evidence Original source ↗ No field-specific location recorded Version: 2409.06744v1 Retrieved: 2026-09-16T21:07:13.231727+00:00	not individually reviewed No individual claim review recorded Audit details Field: `description` Source artifact SHA-256: `4334d636223ad42bfb9ae68aae03f5a255c29ba1cebe7b8f9588fb3b9b5453b2` Hash scope: Exact retrieved primary paper artifact bytes. Inspected artifact
name Co-generated structures and sequences at fixed lengths (ProteinBench split) Context-only references	proteinbench primary benchmark evidence Original source ↗ No field-specific location recorded Version: 2409.06744v1 Retrieved: 2026-09-16T21:07:13.231727+00:00	not individually reviewed No individual claim review recorded Audit details Field: `name` Source artifact SHA-256: `4334d636223ad42bfb9ae68aae03f5a255c29ba1cebe7b8f9588fb3b9b5453b2` Hash scope: Exact retrieved primary paper artifact bytes. Inspected artifact

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

Sources and history

Release 2026-09-17-134cd1815de8 · Record review: source checked

1 source records and release history

proteinbench primary benchmark evidence · Original source · 2409.06744v1

Download this release