Configuration

FrameDiff(latest)

Protein model evaluated by the ProteinBench authors under their harness.

Overview

Protein model evaluated by the ProteinBench authors under their harness.

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

Evaluations and results

Release 2026-09-17-134cd1815de8 · 20 evaluations · 20 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
FrameDiff(latest) on ProteinBench BB-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust. Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-100-DIVERSITY-MAX-CLUST: length 100, Diversity Max Clust.Dataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.11 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 100, Diversity Max Clust.↑) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-100-DIVERSITY-PAIRWISE-TM: length 100, Diversity pairwise TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-100-DIVERSITY-PAIRWISE-TM: length 100, Diversity pairwise TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.52 ± 0.05 pairwise-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.05 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 100, Diversity pairwise TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-100-NOVELTY-MAX-TM: length 100, Novelty Max TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.71 ± 0.08 max-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.08 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 100, Novelty Max TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-100-QUALITY-SCRMSD: length 100, Quality scRMSD Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-100-QUALITY-SCRMSD: length 100, Quality scRMSDDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
1.23 ± 1.02 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 1.02 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 100, Quality scRMSD ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-100-QUALITY-SCTM: length 100, Quality scTM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-100-QUALITY-SCTM: length 100, Quality scTMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.90 ± 0.08 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.08 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 100, Quality scTM ↑) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust. Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-300-DIVERSITY-MAX-CLUST: length 300, Diversity Max Clust.Dataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.21 max-clust Unit: score · Direction: higher	Uncertainty: Not reported Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 300, Diversity Max Clust. ↑) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-300-DIVERSITY-PAIRWISE-TM: length 300, Diversity pairwise TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-300-DIVERSITY-PAIRWISE-TM: length 300, Diversity pairwise TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.48 ± 0.04 pairwise-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.04 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 300, Diversity pairwise TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-300-NOVELTY-MAX-TM: length 300, Novelty Max TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.69 ± 0.00 max-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.00 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 300, Novelty Max TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-300-QUALITY-SCRMSD: length 300, Quality scRMSD Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-300-QUALITY-SCRMSD: length 300, Quality scRMSDDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
2.73 ± 2.69 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 2.69 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 300, Quality scRMSD ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-300-QUALITY-SCTM: length 300, Quality scTM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-300-QUALITY-SCTM: length 300, Quality scTMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.87 ± 0.12 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.12 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 300, Quality scTM ↑) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-50-DIVERSITY-MAX-CLUST: length 50, Diversity Max Clust. Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-50-DIVERSITY-MAX-CLUST: length 50, Diversity Max Clust.Dataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median 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 3, row(FrameDiff(latest)), column(length 50, Diversity Max Clust. ↑) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-50-DIVERSITY-PAIRWISE-TM: length 50, Diversity pairwise TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-50-DIVERSITY-PAIRWISE-TM: length 50, Diversity pairwise TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.35 ± 0.02 pairwise-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.02 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 50, Diversity pairwise TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-50-NOVELTY-MAX-TM: length 50, Novelty Max TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-50-NOVELTY-MAX-TM: length 50, Novelty Max TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.67 ± 0.01 max-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.01 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 50, Novelty Max TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-50-QUALITY-SCRMSD: length 50, Quality scRMSD Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-50-QUALITY-SCRMSD: length 50, Quality scRMSDDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
1.00 ± 1.27 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 1.27 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 50, Quality scRMSD ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-50-QUALITY-SCTM: length 50, Quality scTM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-50-QUALITY-SCTM: length 50, Quality scTMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.85 ± 0.13 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.13 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 50, Quality scTM ↑) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-500-DIVERSITY-MAX-CLUST: length 500, Diversity Max Clust. Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-500-DIVERSITY-MAX-CLUST: length 500, Diversity Max Clust.Dataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median 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 3, row(FrameDiff(latest)), column(length 500, Diversity Max Clust.↑) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-500-DIVERSITY-PAIRWISE-TM: length 500, Diversity pairwise TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-500-DIVERSITY-PAIRWISE-TM: length 500, Diversity pairwise TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.40 ± 0.06 pairwise-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.06 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 500, Diversity pairwise TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-500-NOVELTY-MAX-TM: length 500, Novelty Max TM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-500-NOVELTY-MAX-TM: length 500, Novelty Max TMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.58 ± 0.03 max-tm Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 0.03 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 500, Novelty Max TM ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-500-QUALITY-SCRMSD: length 500, Quality scRMSD Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-500-QUALITY-SCRMSD: length 500, Quality scRMSDDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
9.52 ± 18.19 scrmsd Unit: score · Direction: lower	Uncertainty: type: standard deviation; value: 18.19 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 500, Quality scRMSD ↓) Source checking is not independent reproduction.
FrameDiff(latest) on ProteinBench BB-LENGTH-500-QUALITY-SCTM: length 500, Quality scTM Configuration: FrameDiff(latest)Task: ProteinBench BB-LENGTH-500-QUALITY-SCTM: length 500, Quality scTMDataset subset: Designed backbones at fixed lengths (ProteinBench split) Unconditional backbone design at a fixed sequence length. Values are the median over repeated runs. Author-reported evaluation · Evaluation metadata: source checked
0.63 ± 0.24 sctm Unit: score · Direction: higher	Uncertainty: type: standard deviation; value: 0.24 Scored: Not reported · Eligible: Not reported	source checkedproteinbench primary benchmark evidence · Table 3, row(FrameDiff(latest)), column(length 500, Quality scTM ↑) 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.

0 evidence rows matching the loaded filters