GPT-OSS-120B
Open-weight reasoning model from OpenAI with a mixture-of-experts Transformer architecture, trained by large-scale distillation and reinforcement learning.
Results are recorded on linked configurations and versions of this model. Their settings and evaluations are kept separate below.
Results on linked records
This model has no results of its own in this release. 3 linked records have results. They are not assigned to the underlying model: each record keeps its own settings and evaluations, and its scores are on its own page.
- GPT-OSS-120B (AssayBench) Configuration · configuration of this model · 3 evaluations · 9 results
- GPT-OSS-120B (SFT) (AssayBench) Configuration · configuration of this model · 3 evaluations · 9 results
- GPT-OSS-120B (SFT + GRPO) (AssayBench) Configuration · configuration of this model · 3 evaluations · 9 results
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.
Use this model
How it works, versions and access
Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this model. Consult the linked sources and evaluation records below.
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.
2 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Description Open-weight reasoning model from OpenAI with a mixture-of-experts Transformer architecture, trained by large-scale distillation and reinforcement learning. Individual claims | Abstract: gpt-oss-120b & gpt-oss-20b Model Card (arXiv:2508.10925) qa-descriptions-20261011-b-source-gpt-oss: abstract Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: arXiv:2508.10925 (title and abstract) | source checked ["automated-source-review","ai-assisted-source-review"] · 2026-10-11 Audit detailsApproved. Checked the new sources' licences against the arXiv abstract pages and Europe PMC, the 7 scFoundation pipeline descriptions against their abstracts, the language-model descriptions against their model cards and reports, and the TAPE architecture figures against the TAPE paper; all correct. The catalogue entries reuse reviewed profile summaries. Field: Claim: metadata-correction-821a80c9eb3bcee03b35 Source artifact SHA-256: Hash scope: SHA-256 of one retrieval of an arXiv API Atom response for this identifier. The response is not guaranteed byte-stable; the title and summary fields are the evidence. |
| Description Open-weight reasoning model from OpenAI with a mixture-of-experts Transformer architecture, trained by large-scale distillation and reinforcement learning. Individual claims | AssayBench: An Assay-Level Virtual Cell Benchmark for LLMs and Agents qa-descriptions-20261011-b-source-gpt-oss: abstract Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: arXiv:2605.10876 version 1, posted 2026-05-11; not peer reviewed | source checked ["automated-source-review","ai-assisted-source-review"] · 2026-10-11 Audit detailsApproved. Checked the new sources' licences against the arXiv abstract pages and Europe PMC, the 7 scFoundation pipeline descriptions against their abstracts, the language-model descriptions against their model cards and reports, and the TAPE architecture figures against the TAPE paper; all correct. The catalogue entries reuse reviewed profile summaries. Field: Claim: metadata-correction-821a80c9eb3bcee03b35 Source artifact SHA-256: Hash scope: pdftotext -layout text layer, parsed by extract/extract_target_validation.py |
Sources and history
Release 2026-11-02-a7f03c9e55a5 · Record review: source checked
2 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
- Abstract: gpt-oss-120b & gpt-oss-20b Model Card (arXiv:2508.10925) · Original source · arXiv:2508.10925 (title and abstract)
Technical metadata and extraction receipts
Stable ID: tgtval-20261009-model-gpt-oss-120b
- areas
- cells-tissues
- contexts
- research
- entity level
- checkpoint
- reported name
- GPT-OSS-120B
- source locator
- AssayBench Tables 2-3 row labels; section 4.1
- missing metadata
- checkpoint revision: reason: unreported; training data: reason: unreported; version: reason: unreported
Related records
- subject: GPT-OSS-120B: description
- configuration of: GPT-OSS-120B (AssayBench)
- configuration of: GPT-OSS-120B (SFT) (AssayBench)
- configuration of: GPT-OSS-120B (SFT + GRPO) (AssayBench)