Internal adversarially-designed factuality evaluations
Models answer adversarial questions in nine categories such as history, math and code. Scored on agreement with human ideal responses.
In-house OpenAI evaluation. Results appear as a chart by category.
- Made by
- OpenAI
- Released
- March 2023
- Kind
- Hallucination rates
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- Kept private
- Strongest model tested
- GPT-6 Astra #3 on the capability index
- Last in a launch post
- September 2026
- Last model added
- September 2026
GPT-6 Luna leads with 7.60%
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.
- 1 GPT-6 Luna 7.60%
- 2 GPT-6 Astra 4.20%
Source: Labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Luna | 7.60% | Sep 2026 | Lab model card | |
| 2 | GPT-6 Astra | 4.20% | Sep 2026 | Lab model card |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Labs’ model cards
One lab put it in a launch post in the last six months
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| Sep 22, 2026 | GPT-6 Luna | 7.6 | Internal factuality evaluation, GPT-6 Luna, factual error rate |
2 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 |
|---|---|---|---|---|
| Sep 3, 2026 | GPT-6 Astra system card | 4.2 | Internal hallucination benchmark, lower is better | Hallucination |
| Mar 14, 2023 | GPT-4 technical report | Internal factual eval by category, nine adversarially-designed evaluations, chart without printed numbers, 19 points above latest GPT-3.5 (Figure 6) | 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% - 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. GPT-6 Astra 4.20%
- AA-Omniscience v1.1 In use Models answer 3,000 knowledge questions in six domains without tools and may abstain. The index adds correct answers and subtracts wrong ones.
Inkling 2 Small 16.1 - 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%