100Q-Hard (Anthropic)
Claude Mythos Preview 60.1% leads since Apr 2026 Only in system cards
Models answer hard, obscure factual questions. Scored on correct, incorrect and declined answers.
In-house Anthropic evaluation. The tasks are not public.
Made by Anthropic Released Sep 2025 Short factual questions Tasks kept private Last in a system card May 2026
Last model added May 2026
- Level
- Expert Specialists with years of training, often over hours of work
- 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.
7
Models scored in lab cards
9
System cards report it, from 1 lab
Claude Mythos Preview leads with 60.1% among models released by May 2026
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 May 2026, 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 15 more
- 1 Claude Mythos Preview 60.1%
- 2 Claude 3 Opus 46.5%
- 3 Claude Opus 4.5 45.9%
- 4 Claude Sonnet 4.5 38.3%
- 5 Claude Opus 4.6 21.5%
- 6 Claude Opus 4.8 20%
- 7 Claude Opus 4.7 17.1%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Mythos Preview | 60.1% | Apr 2026 | Model card | |
| 2 | Claude 3 Opus | 46.5% | Mar 2024 | Model card | |
| 3 | Claude Opus 4.5 | 45.9% | Nov 2025 | Model card | |
| 4 | Claude Sonnet 4.5 | 38.3% | Sep 2025 | Model card | |
| 5 | Claude Opus 4.6 | 21.5% | Feb 2026 | Model card | |
| 6 | Claude Opus 4.8 | 20% | May 2026 | Model card | |
| 7 | Claude Opus 4.7 | 17.1% | Apr 2026 | Model card |
2 models have held number one since its release in Sep 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Anthropic
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Reported by the lab in its model card
Best so far
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 |
|---|---|---|---|---|
| 28 May 2026 | Claude Opus 4.8 system card | 20 | 100Q-Hard net score 0.20 (correct minus incorrect), closed book, printed on chart (6.3.3.1, Figure 6.3.3.1.B) | Hallucination |
| 16 Apr 2026 | Claude Opus 4.7 system card | 17.1 | 100Q-Hard net score (correct minus incorrect), closed book, printed on chart. Correct 47.2, unsure 22.6, incorrect 30.2 (6.3.3.1, Figures 6.3.3.1.A and B) | Hallucination |
| 7 Apr 2026 | Claude Mythos Preview system card | 60.1 | 100Q-Hard correct rate, no tools, value printed on chart; unsure 12.9, incorrect 27.0, net score not printed (4.3.3.1, Figure 4.3.3.1.A) | Hallucination |
| 5 Feb 2026 | Claude Opus 4.6 system card | 21.5 | 100Q-Hard net score (correct minus incorrect), extended thinking (16k), value printed on chart; correct rate 43.9, net 9.8 with thinking plus effort, net 20.0 without thinking (Figure 4.2.A) | Hallucination |
| 24 Nov 2025 | Claude Opus 4.5 system card | 45.9 | 100Q-Hard correct rate, 16k extended thinking (Figure 4.1.A; incorrect 34.0%, uncertain 20.1%) | Hallucination |
| 29 Sep 2025 | Claude Sonnet 4.5 system card | 38.3 | 100Q-Hard correct answer rate, with thinking, no tools, graded by Claude Sonnet 4. Incorrect 37.3%, uncertain 24.4%. Without thinking correct 32.9%, incorrect 44.0% (Figure 3.2.A) | Hallucination |
| 4 Mar 2024 | Claude 3 Sonnet model card | 100Q Hard, correct, incorrect and unsure shares, chart without printed values (Figure 11) | Hallucination | |
| 4 Mar 2024 | Claude 3 Opus model card | 46.5 | 100Q Hard, % correct, stated in text. Incorrect and unsure shares in a chart without printed values (5.7, Figure 11) | Hallucination |
| 4 Mar 2024 | Claude 3 Haiku model card | 100Q Hard, correct, incorrect and unsure shares, chart without printed values (Figure 11) | 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%