BlogLLM Benchmark Scores as Published, Side by Side
ChatGPT answers without errors (internal): Results as Published
Every result the catalog holds for ChatGPT answers without errors (internal): 2 models from 1 developer, read from 1 page. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area General Knowledge and ReasoningMaintainer OpenAIData version 2026103001, as of 2026-10-30
An internal OpenAI factuality evaluation on a set of de-identified ChatGPT queries. Each response is checked for errors by other models, and the result is the share of responses without any error, with and without a search tool. Claim-level error rates are far lower than response-level rates because most responses contain many claims.
Not checked against a maintainer's page.
ChatGPT answers without errors (no search)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.2OpenAI, released 2025-12-11 | 88% | OpenAI, Introducing GPT-5.2checked 2026-10-15 | Effort: xhigh; Tools: none |
| GPT-5.1OpenAI, released 2025-11-12 | 87.3% | OpenAI, Introducing GPT-5.2checked 2026-10-15 | Effort: high; Tools: none |
ChatGPT answers without errors (with search)
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.2OpenAI, released 2025-12-11 | 93.9% | OpenAI, Introducing GPT-5.2checked 2026-10-15 | Effort: xhigh; Tools: search |
| GPT-5.1OpenAI, released 2025-11-12 | 91.2% | OpenAI, Introducing GPT-5.2checked 2026-10-15 | Effort: high; Tools: search |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.