Configuration

GPT-OSS-120B (SFT + GRPO) (AssayBench)

Configuration as run in the cited comparison.

Overview

Configuration as run in the cited comparison.

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

Evaluations and results

3 evaluations · 9 results. Different protocols are not a single leaderboard.

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: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench post-cutoff cohort of human CRISPR screens	0.101 adjusted-ndcg-at-100 unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench post-cutoff cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'LaTest', column 'AnDCG@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench post-cutoff cohort of human CRISPR screens	NA directional-false-discovery-rate-at-100 fraction · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench post-cutoff cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'LaTest', column 'dFDR@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench post-cutoff cohort of human CRISPR screens	0.253 precision-at-100 fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench post-cutoff cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'LaTest', column 'Precision@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench test cohort of human CRISPR screens	0.129 adjusted-ndcg-at-100 unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench test cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'test', column 'AnDCG@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench test cohort of human CRISPR screens	0.0252 directional-false-discovery-rate-at-100 fraction · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench test cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'test', column 'dFDR@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench test cohort of human CRISPR screens	0.17 precision-at-100 fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench test cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'test', column 'Precision@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench validation cohort of human CRISPR screens	0.136 adjusted-ndcg-at-100 unitless · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench validation cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'val', column 'AnDCG@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench validation cohort of human CRISPR screens	0.0165 directional-false-discovery-rate-at-100 fraction · lower Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench validation cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'val', column 'dFDR@100'
Configuration: GPT-OSS-120B (SFT + GRPO) (AssayBench)	Protocol: AssayBench: rank 100 candidate genes for a described CRISPR screen Dataset: AssayBench validation cohort of human CRISPR screens	0.231 precision-at-100 fraction · higher Uncertainty: Not reported by the source Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · Source checked Methods, coverage and source GPT-OSS-120B (SFT + GRPO) on the AssayBench validation cohort tgtval-20261009-protocol-debrouwer2026-screen-gene-ranking Aggregation: Not reported AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Table 3, row 'GPT-OSS-120B (SFT + GRPO)' cohort 'val', column 'Precision@100'

Source checking is not independent reproduction. Release 2026-10-10-6e93f504adfc.

Use this model

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 2026-10-10-6e93f504adfc
Property and statement	Original source and location	Review and provenance

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

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

Sources and history

Release 2026-10-10-6e93f504adfc · Record review: source checked

1 source records and release history

AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents · Original source · arXiv:2605.10876 version 1, posted 2026-05-11; not peer reviewed

Download this release (gzip)

Technical metadata and extraction receipts

Stable ID: tgtval-20261009-config-debrouwer2026-gpt-oss-120b-sft-grpo

areas: cells-tissues
contexts: research
method types: foundation_model
reported name: GPT-OSS-120B (SFT + GRPO)
foundation model eligible: true
source locator: Tables 2-3 row labels; section 4
missing metadata: version: reason: unreported; note: No release or commit is printed for this configuration
parameters: supervised fine-tuning then GRPO reinforcement learning

Related records

configuration of: GPT-OSS-120B
system: GPT-OSS-120B (SFT + GRPO) on the AssayBench post-cutoff cohort
system: GPT-OSS-120B (SFT + GRPO) on the AssayBench test cohort
system: GPT-OSS-120B (SFT + GRPO) on the AssayBench validation cohort

Suggest a correction