GenomeOcean-500M, FP8 (E4M3) static weights, PyTorch with FBGEMM (Yu et al. 2026)
Protocol A: static FP8 E4M3 weight quantization with FBGEMM kernels; output embeddings in BF16; no calibration fine-tuning.
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.
Evaluations and results
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% |
All 4 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
Source checking is not independent reproduction. Release 2026-10-22-1b7cf027e4f6.
Use this configuration
How it works, versions and access
Strengths, limitations and unresolved questions
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
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
Sources and history
Release 2026-10-22-1b7cf027e4f6 · Record review: source checked
1 source record and release history
- FP8 Inference in Genomic Foundation Models: Theoretical vs. Realized Speedups on GenomeOcean · Original source · bioRxiv preprint version 1, posted 2026-08-14 (PDF footer); bioRxiv API date 2026-08-18
Technical metadata and extraction receipts
Stable ID: model-execution-20261011-config-yu2026-genomeocean-500m-fp8-weights-embedding
- areas
- dna-genomes
- contexts
- research
- method types
- foundation_model
- reported name
- 500M FP8 (E4M3) static weights
- foundation model eligible
- true
- version
- GenomeOcean-500M
- parameters
- 500M parameters
- access reuse
- Checkpoint release not stated in this source; benchmarking code at https://github.com/jgi-genomeocean/genomeocean_efficiency (not pinned)
- hardware
- description: One NVIDIA H100 SXM5 GPU, 80 GB HBM3 (Section 2.2)
- protocol
- Protocol A: static FP8 E4M3 weight quantization with FBGEMM kernels; output embeddings in BF16; no calibration fine-tuning.
- source locator
- Sections 2.1-2.4; Table 1; Table 2
Related records
- configuration of: GenomeOcean
- system: GenomeOcean-500M FP8 (E4M3) embedding extraction (Yu et al. 2026)