ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation
Web UI across four paid accounts; memory disabled, fresh chats; some concurrent tabs; underlying model not pinnable.
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.
Model Capabilities & Hardware Requirements
Deployment facts recorded for ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation. Fields the record does not carry are marked not recorded.
Evaluations and results
1 evaluation · 54 results. Different protocols are not a single leaderboard.
Protocol: CIViC evidence-direction retrieval, March 2026 · Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026
Sorted by Accuracy (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 | Accuracy | F1 (class: diagnostic negative) | F1 (class: diagnostic positive) | F1 (class: functional dominant negative) | F1 (class: functional gain of function) | F1 (class: functional loss of function) | F1 (class: functional unaltered function) | F1 (class: oncogenic oncogenicity) | F1 (class: predictive reduced sensitivity) | F1 (class: predictive resistance) | F1 (class: predictive sensitivity) | F1 (class: predisposing n/a) | F1 (class: predisposing predisposition) | F1 (class: predisposing uncertain significance) | F1 (class: prognostic better outcome) | F1 (class: prognostic poor outcome) | Micro F1 | Micro F1 (evidence type: diagnostic) | Micro F1 (evidence type: functional) | Micro F1 (evidence type: oncogenic) | Micro F1 (evidence type: predictive) | Micro F1 (evidence type: predisposing) | Micro F1 (evidence type: prognostic) | Precision (class: diagnostic negative) | Precision (class: diagnostic positive) | Precision (class: functional dominant negative) | Precision (class: functional gain of function) | Precision (class: functional loss of function) | Precision (class: functional unaltered function) | Precision (class: oncogenic oncogenicity) | Precision (class: predictive reduced sensitivity) | Precision (class: predictive resistance) | Precision (class: predictive sensitivity) | Precision (class: predisposing n/a) | Precision (class: predisposing predisposition) | Precision (class: predisposing uncertain significance) | Precision (class: prognostic better outcome) | Precision (class: prognostic poor outcome) | Recall (class: diagnostic negative) | Recall (class: diagnostic positive) | Recall (class: functional dominant negative) | Recall (class: functional gain of function) | Recall (class: functional loss of function) | Recall (class: functional unaltered function) | Recall (class: oncogenic oncogenicity) | Recall (class: predictive reduced sensitivity) | Recall (class: predictive resistance) | Recall (class: predictive sensitivity) | Recall (class: predisposing n/a) | Recall (class: predisposing predisposition) | Recall (class: predisposing uncertain significance) | Recall (class: prognostic better outcome) | Recall (class: prognostic poor outcome) | Runtime (mean per query) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluationEvaluation | 0.83 | 0 | 0.94 | 1 | 0.67 | 0.92 | 1 | 0.96 | 1 | 0.94 | 0.88 | 0.88 | 0.96 | 0.92 | 1 | 0.89 | 0.91 | 0.84 | 0.88 | 0.96 | 0.9 | 0.93 | 0.95 | 0 | 0.89 | 1 | 0.75 | 0.86 | 1 | 0.92 | 1 | 1 | 1 | 1 | 0.93 | 1 | 1 | 1 | 0 | 1 | 1 | 0.6 | 1 | 1 | 1 | 1 | 0.88 | 0.79 | 0.78 | 1 | 0.86 | 1 | 0.8 | 425 |
All 54 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.83 Accuracy fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP: integrating large language models with the Clinical Interpretations of Variants in Cancer · Fig1c Overall Accuracy; Section 3 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.84 Micro F1 (evidence type: diagnostic) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP: integrating large language models with the Clinical Interpretations of Variants in Cancer · Figure 1c, diagnostic row, agent column |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0 F1 (class: diagnostic negative) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Diagnostic / Negative, f1 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0 Precision (class: diagnostic negative) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Diagnostic / Negative, precision |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0 Recall (class: diagnostic negative) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Diagnostic / Negative, recall |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.94 F1 (class: diagnostic positive) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Diagnostic / Positive, f1 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.89 Precision (class: diagnostic positive) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Diagnostic / Positive, precision |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 Recall (class: diagnostic positive) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Diagnostic / Positive, recall |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 F1 (class: functional dominant negative) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Dominant Negative, f1 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 Precision (class: functional dominant negative) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Dominant Negative, precision |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 Recall (class: functional dominant negative) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Dominant Negative, recall |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.88 Micro F1 (evidence type: functional) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP: integrating large language models with the Clinical Interpretations of Variants in Cancer · Figure 1c, functional row, agent column |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.67 F1 (class: functional gain of function) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Gain of Function, f1 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.75 Precision (class: functional gain of function) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Gain of Function, precision |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.6 Recall (class: functional gain of function) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Gain of Function, recall |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.92 F1 (class: functional loss of function) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Loss of Function, f1 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.86 Precision (class: functional loss of function) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Loss of Function, precision |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 Recall (class: functional loss of function) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Loss of Function, recall |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 F1 (class: functional unaltered function) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Unaltered Function, f1 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 Precision (class: functional unaltered function) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Unaltered Function, precision |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 1 Recall (class: functional unaltered function) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Functional / Unaltered Function, recall |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.91 Micro F1 unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP: integrating large language models with the Clinical Interpretations of Variants in Cancer · Fig1c Micro F1; Section 3 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.96 Micro F1 (evidence type: oncogenic) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP: integrating large language models with the Clinical Interpretations of Variants in Cancer · Figure 1c, oncogenic row, agent column |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.96 F1 (class: oncogenic oncogenicity) unitless · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Oncogenic / Oncogenicity, f1 |
| Configuration: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation | Protocol: CIViC evidence-direction retrieval, March 2026 Dataset: CIViC 100 molecular-profile/disease/therapy triplets, March 2026 | 0.92 Precision (class: oncogenic oncogenicity) fraction · higher Uncertainty: Not reported by the source: not reported for this endpoint Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceNot reported Aggregation: Not reported CIViC MCP supplementary methods and complete Tables 1–3 · Supplementary Table 3, Oncogenic / Oncogenicity, precision |
Source checking is not independent reproduction. Release 2026-11-21-c8bea25e5abb.
Use this configuration
How it works, versions and access
Related profile: CIViC fixed-schema evidence retrieval evaluation workflow. This page retains the exact record and its evaluation context.
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 |
|---|---|---|
Facets: method types1 values
| CIViC MCP supplementary methods and complete Tables 1–3 Record names and descriptions: GPT-5, ChatGPT Agent Mode and Qwen3-Reranker-8B are large pretrained language models used without task training Version: vbag209_supplementary_data.docx, final article supplement | conflicting claim ["automated-source-review"] · 2026-10-12 Audit detailsCorrect the method type from none to foundation_model (use-case baseline review 2026-10-12, table 2). The review marks these as the proposer's, not source-checked. Evidence from docs/reviews/use-cases/baselines-2026-10-12.md. Independent review 2026-10-12: approved. See docs/reviews/2026-10-12-method-type-corrections.md. Conflicting source claim: Field: Claim: metadata-correction-tags-20261012-190eda97d02de0f7855e Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Relationship: configuration of uc-clinical-20260930-pipeline-civic-retrieval Individual claims | CIViC MCP supplementary methods and complete Tables 1–3 Supplementary implementation and Agent Mode evaluation methods Version: vbag209_supplementary_data.docx, final article supplement | source checked ["automated-source-review"] · 2026-09-30 Audit detailsTranscription from inspected primary source. No execution, independent replication or human scientific review. Field: Claim: uc-clinical-20260930-association-civic-agent-mode Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
Release 2026-11-21-c8bea25e5abb · Record review: source checked
1 source record and release history
- CIViC MCP supplementary methods and complete Tables 1–3 · Original source · vbag209_supplementary_data.docx, final article supplement
Technical metadata and record identifiers
Technical metadata for this record is tracked under stable ID uc-clinical-20260930-civic-agent-mode.
Related records
- configuration of: CIViC fixed-schema evidence retrieval evaluation workflow
- subject: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation: facets.method_types
- subject: Configuration identity: ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation
- implemented by: Browser Agent Mode retrieval baseline
- system: CIViC agent retrieval
How to cite
ChatGPT Agent Mode, 2–7 March 2026 CIViC evaluation. Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/uc-clinical-20260930-civic-agent-mode/
Release 2026-11-21-c8bea25e5abb. Rankings and values change between releases, so keep the release id in the citation. Cite the original publication as well for any value you reuse: each result links its source.
The compilation and annotations made by Rewire in this release are offered under CC BY 4.0; values and text taken from cited sources stay under the terms of those sources and are not covered by this licence. Data licence