Configuration

NTv3 100M training run (Boshar et al. 2025)

Training of NTv3 100M through all stages reported in Table 5.

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 · 5 results. Different protocols are not a single leaderboard.

Protocol: NTv3 H100 GPU-hours per training stage (Boshar et al. 2025 Table 5) · Dataset: NTv3 training data, all stages (Boshar et al. 2025)

Sorted by Accelerator time (Training stage 'Length ext. 1-4' (200B tokens)) (lower 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	Accelerator time (Training stage 'Length ext. 1-4' (200B tokens))	Accelerator time (Training stage 'Length ext. 5-7' (150B tokens))	Accelerator time (Training stage 'Post-training 1' (1.2T tokens))	Accelerator time (Training stage 'Post-training 2' (54B tokens))	Accelerator time (Training stage 'Pre-training 8 kb' (10.5T tokens))
NTv3 100M training run (Boshar et al. 2025)Evaluation	40	48	1,536	192	3,000

All 5 result rows with coverage, uncertainty and sources

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: NTv3 100M training run (Boshar et al. 2025)	Protocol: NTv3 H100 GPU-hours per training stage (Boshar et al. 2025 Table 5) Dataset: NTv3 training data, all stages (Boshar et al. 2025)	40 Accelerator time (Training stage 'Length ext. 1-4' (200B tokens)) hour · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NTv3 100M training GPU-hours (Boshar et al. 2025 Table 5) gfm-guide-20261014-protocol-ntv3-training-gpu-hours Aggregation: Not reported A foundational model for joint sequence-function multi-species modeling at scale for long-range genomic prediction (NTv3), bioRxiv v1 · Table 5 (page 34), row '100M', column 'Length ext. 1-4'
Configuration: NTv3 100M training run (Boshar et al. 2025)	Protocol: NTv3 H100 GPU-hours per training stage (Boshar et al. 2025 Table 5) Dataset: NTv3 training data, all stages (Boshar et al. 2025)	48 Accelerator time (Training stage 'Length ext. 5-7' (150B tokens)) hour · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NTv3 100M training GPU-hours (Boshar et al. 2025 Table 5) gfm-guide-20261014-protocol-ntv3-training-gpu-hours Aggregation: Not reported A foundational model for joint sequence-function multi-species modeling at scale for long-range genomic prediction (NTv3), bioRxiv v1 · Table 5 (page 34), row '100M', column 'Length ext. 5-7'
Configuration: NTv3 100M training run (Boshar et al. 2025)	Protocol: NTv3 H100 GPU-hours per training stage (Boshar et al. 2025 Table 5) Dataset: NTv3 training data, all stages (Boshar et al. 2025)	1,536 Accelerator time (Training stage 'Post-training 1' (1.2T tokens)) hour · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NTv3 100M training GPU-hours (Boshar et al. 2025 Table 5) gfm-guide-20261014-protocol-ntv3-training-gpu-hours Aggregation: Not reported A foundational model for joint sequence-function multi-species modeling at scale for long-range genomic prediction (NTv3), bioRxiv v1 · Table 5 (page 34), row '100M', column 'Post-training 1'
Configuration: NTv3 100M training run (Boshar et al. 2025)	Protocol: NTv3 H100 GPU-hours per training stage (Boshar et al. 2025 Table 5) Dataset: NTv3 training data, all stages (Boshar et al. 2025)	192 Accelerator time (Training stage 'Post-training 2' (54B tokens)) hour · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NTv3 100M training GPU-hours (Boshar et al. 2025 Table 5) gfm-guide-20261014-protocol-ntv3-training-gpu-hours Aggregation: Not reported A foundational model for joint sequence-function multi-species modeling at scale for long-range genomic prediction (NTv3), bioRxiv v1 · Table 5 (page 34), row '100M', column 'Post-training 2'
Configuration: NTv3 100M training run (Boshar et al. 2025)	Protocol: NTv3 H100 GPU-hours per training stage (Boshar et al. 2025 Table 5) Dataset: NTv3 training data, all stages (Boshar et al. 2025)	3,000 Accelerator time (Training stage 'Pre-training 8 kb' (10.5T tokens)) hour · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source NTv3 100M training GPU-hours (Boshar et al. 2025 Table 5) gfm-guide-20261014-protocol-ntv3-training-gpu-hours Aggregation: Not reported A foundational model for joint sequence-function multi-species modeling at scale for long-range genomic prediction (NTv3), bioRxiv v1 · Table 5 (page 34), row '100M', column 'Pre-training 8 kb'

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

Claims, original sources and review scope · Release 2026-10-22-1b7cf027e4f6
Property and statement	Original source and location	Review and provenance

No evidence rows match these filters. Choose another scope or clear the search.

How to read the evidence

Sources and history

Release 2026-10-22-1b7cf027e4f6 · Record review: source checked

1 source record and release history

A foundational model for joint sequence-function multi-species modeling at scale for long-range genomic prediction (NTv3), bioRxiv v1 · Original source · bioRxiv v1, posted 2025-12-25 (dated 2025-12-22 on the title page)

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: gfm-guide-20261014-config-ntv3-100m-training

areas: dna-genomes
contexts: research
method types: foundation_model
reported name: NTv3 100M
hardware: description: Machines with 8 NVIDIA H100 80GB GPUs connected via NVLink, nodes connected with RoCE; some runs on Cloud TPU v6 (section 4.4.4)
source locator: Table 5; section 4.4.4
missing metadata: version: reason: unreported; note: No checkpoint revision named in the table

Related records

configuration of: Nucleotide Transformer v3
system: NTv3 100M training GPU-hours (Boshar et al. 2025 Table 5)

Suggest a correction