GPT-6.1 Sol · OpenAI difficult-prompt factuality (internal; version unspecified)
Factuality · provider-reported by OpenAI. Evaluation snapshot: unknown. Test date: unknown.
Low reasoning effort. Share of answers containing at least one factual error, on de-identified conversations where users flagged an earlier error. Not representative of typical use. Agent/tools, attempts, sample size and exact snapshot not disclosed.
Exact public release label matched to catalog ID; evaluation snapshot not disclosed. Not a guarantee for future revisions of the API alias.
Original report · Methodology / system card · Source checked 2026-11-12 (not the test date).