Published inference benchmark; transcribed, not reproduced.
1 evaluation · 4 results. Different protocols are not a single leaderboard.
Protocol: GenomeOcean embedding extraction throughput, peak VRAM and binning fidelity, BF16 versus FP8 weights (Yu et al. 2026 Table 3) · Dataset: GTDB genome fragments for metagenomic binning, 20 of a pre-sampled pool of 50 bacterial genomes per trial (Yu et al. 2026)
Sorted by Adjusted rand index (DBSCAN clustering of embeddings against GTDB species labels) (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 configuration | Adjusted rand index (DBSCAN clustering of embeddings against GTDB species labels) | Inference throughput (Embedding extraction; input tokens per second) | Peak memory (Embedding extraction; peak GPU memory allocated) | Throughput change (Embedding extraction; relative to the BF16 baseline of the same model scale) |
|---|---|---|---|---|
| GenomeOcean-500M, FP8 (E4M3) static weights, PyTorch with FBGEMM (Yu et al. 2026)Evaluation | 0.3273 ± 0.0508 | 204,973 ± 777 | 46.7 | −11.1% |
Applied filters: All linked evaluations
Source checking is not independent reproduction. Release 2026-10-22-1b7cf027e4f6.
0 of 4 readiness checks met. These checks assess whether the evidence supports a reproducible investigation; a source-checked score alone does not meet them.
Release 2026-10-22-1b7cf027e4f6 · Evidence verified: Not verified
Exact outcomes, predictions, identifiers and evaluator are connected.
Missing or unresolved evidence
Verified: Not verified
Replay evidence includes annotations and an assessment of dependence. Unknown independence permits descriptive analysis only.
Missing or unresolved evidence
Verified: Not verified
A pinned recipe describes the inputs, environment and resource requirements.
Missing or unresolved evidence
Verified: Not verified
Separate data and exposure records support an independent test.
Missing or unresolved evidence
Verified: Not verified
Readiness describes the evidence in this release. Availability on your computer is checked separately when an investigation runs. Existing data exposure can prevent independent validation even when files are available.
No verified artifact manifest is connected to this record yet. The gaps above identify what is needed before analysis can begin.
Metadata review: source checked. Unreported conditions prevent automatic comparisons.
No execution recipe has been verified for this exact configuration and evaluation. A benchmark's general instructions may use different inputs, splits or model settings.
Reproducing this published result requires matching its model configuration, data, split and scorer. Source checking or a successful smoke test does not establish score reproduction.
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
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.
18 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Comparison: adaptation None (pretrained checkpoint, post-training quantization only) Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: aggregation Mean and standard deviation over five trials Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: budget One NVIDIA H100 SXM5 80 GB GPU Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: dataset version Not reported Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | missing or unspecified No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: inputs Tokenized DNA sequence Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: metric implementation Not reported Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | missing or unspecified No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: population GTDB bacterial genomes Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: protocol id model-execution-20261011-protocol-yu2026-genomeocean-embedding-fp8 Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Comparison: split Five trials, seeds 42-46 Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
| Origin author_reported Context-only references | FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean Table 3 row '500M FP8 (E4M3)' Version: bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18 | not individually reviewed No individual claim review recorded author reported Audit detailsField: Source artifact SHA-256: Hash scope: bioRxiv stamps each PDF download, so artifact_sha256 describes one retrieval (a second download differed in 40 bytes). The pdftotext -layout text layer (poppler) of both downloads has SHA-256 bec6e695ce46c4ebd6692e01a1a93167e212dffbb97b6558f1da88b1eb453501. |
Release 2026-10-22-1b7cf027e4f6 · Record review: source checked
Stable ID: model-execution-20261011-eval-yu2026-genomeocean-500m-fp8-weights-embedding