PersonQA
Models answer questions about publicly available facts on people. Scored on accuracy, with hallucination rate reported alongside.
In-house OpenAI evaluation. Results here record accuracy and name the hallucination rate in the text.
- Level
- Everyday Most adults, without training
- Scored
- Graded by a model A language model scores each response, usually against a rubric.
- Answers
- Short answer A number, word or short string matched to a key.
ChatGPT agent leads with 94.3% among models released by Aug 2025
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in Aug 2025, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 ChatGPT agent 94.3%
- 2 OpenAI deep research 86%
- 3 GPT-4.5 78%
- 4 o3 59%
- 5 o1 55%
- 6 o4-mini 36%
- 7 gpt-oss-120b 29.8%
- 8 o3-mini 21.7%
- 9 gpt-oss-20b 15.5%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | ChatGPT agent | 94.3% | Jul 2025 | Model card | |
| 2 | OpenAI deep research | 86% | Feb 2025 | Model card | |
| 3 | GPT-4.5 | 78% | Feb 2025 | Model card | |
| 4 | o3 | 59% | Apr 2025 | Model card | |
| 5 | o1 | 55% | Dec 2024 | Model card | |
| 6 | o4-mini | 36% | Apr 2025 | Model card | |
| 7 | gpt-oss-120b | 29.8% | Aug 2025 | Model card | |
| 8 | o3-mini | 21.7% | Jan 2025 | Model card | |
| 9 | gpt-oss-20b | 15.5% | Aug 2025 | Model card |
3 models have held number one since its release in Dec 2024
Lab claims count from the day the card came out, independent runs from the model’s release.
The best score passed 85% in Feb 2025
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs report it only in system cards
We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark. Some system cards do, below.
9 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 5 Aug 2025 | gpt-oss-20b model card | 15.5 | PersonQA, accuracy, no browsing (hallucination rate 53.2%) (Table 9) | Hallucination |
| 5 Aug 2025 | gpt-oss-120b model card | 29.8 | PersonQA, accuracy, no browsing (hallucination rate 49.1%) (Table 9) | Hallucination |
| 17 Jul 2025 | ChatGPT agent system card | 94.3 | PersonQA, accuracy, hallucination rate 0.043 (Table 4) | Hallucination |
| 16 Apr 2025 | o4-mini system card | 36 | PersonQA, accuracy, hallucination rate 0.48 (Table 4) | Hallucination |
| 16 Apr 2025 | o3 system card | 59 | PersonQA, accuracy, hallucination rate 0.33 (Table 4) | Hallucination |
| 27 Feb 2025 | GPT-4.5 system card | 78 | PersonQA, accuracy, hallucination rate 0.19 (Table 4) | Hallucination |
| 25 Feb 2025 | OpenAI deep research system card | 86 | PersonQA, accuracy 0.86, hallucination rate 0.13 (Table 11) | Hallucination |
| 31 Jan 2025 | o3-mini system card | 21.7 | PersonQA accuracy, hallucination rate 14.8% (Table 3) | Hallucination |
| 5 Dec 2024 | o1 system card | 55 | PersonQA, accuracy (hallucination rate 0.20) (Table 3) | Hallucination |
More in Factuality
- SimpleVQA In use Models answer 2,025 short factual questions about images across 9 tasks and 9 topics. An LLM judge grades answers against fixed references.
Qwen3.7-Plus 81.7% - HallusionBench In use Models answer 1,129 yes or no questions about 346 images, built in pairs to expose language hallucination and visual illusion. Scored by accuracy.
Keye-VL-2.0-30B-A3B 73.5% - SimpleQA Verified In use 1,000 human-verified, de-duplicated short-answer factual questions measuring parametric knowledge.
Gemini 3.1 Pro 77.5% - GaRAGe In use Models write long-form RAG answers to 2,366 questions with human-annotated grounding passages. Scoring checks whether answers use the relevant passages.
Nemotron 3.5 Bolt E4B 40.7% - Kimi Hallu Bench In use Models answer hard prompts drawn from practical work tasks. Fact-checking agents score the share of claims that are hallucinated. Kimi K3.1 Pro 8.30%
- RAGTruth In use A word-level annotated corpus of hallucinations in retrieval-augmented generation answers. It is used to train and evaluate hallucination detection and RAG faithfulness. Nemotron 3.5 Bolt E4B 77.9%