Run detail

j976yztmsq9f35crkvwgr6wcdn85ef9d

Canonical full-suite run for GPT-5.4. Raw response path: artifacts/paid-latest-labs-2026-04-24/full/openai_gpt-5.4/a89cc11f/j976yztmsq9f35crkvwgr6wcdn85ef9d.responses.jsonl.

Completion100%
Parse validity100%
Mean confidence0.83
Average latency1,773 ms
Score statusrenderable

No suppression reasons

Raw evidenceartifacts/paid-latest-labs-2026-04-24/full/openai_gpt-5.4/a89cc11f/j976yztmsq9f35crkvwgr6wcdn85ef9d.responses.jsonl

Response Style

Strong agree26.7%
Agree44.8%
Neutral4.4%
Disagree17.4%
Strong disagree6.7%
Refusal0%
Parse error0%

Traceability

Question signaturefull.ef7b7fd6
Model roster signaturemodels.v1.3.1.98884c4a
Prompt templatept.v1.1.1
Parser versionparser.v1.0.1
Scorer versionsc.v1.2.0
Canonical responses/polibench-paper-v1.0.1/canonical_responses.csv#j976yztmsq9f35crkvwgr6wcdn85ef9d
Duplicate decisions/polibench-paper-v1.0.1/duplicate_resolution.csv#j976yztmsq9f35crkvwgr6wcdn85ef9d