StressEval - Text Knowledge Stress: leaderboard
Metric: Exact-match accuracy (%) on the text knowledge-stress split (468 instances seeded from HotpotQA, answering needs a fact withheld from the context) of Dynamic-OneEval (StressEval instances synthesized by GPT-5.2 from its own failure cases on HotpotQA, ComplexWebQuestions and WikiTableQuestions, gated for answerability and consistency), one unified prompt template for every model; higher is better. Source: arxiv.org. Saturation forecast: Around August 2027. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 54.3 |
| 2 | Seed-1.6 | 32.9 |
| 3 | Claude Sonnet 4.5 | 31.2 |
| 4 | Llama 3.1 70B | 21.4 |
| 5 | QwQ-32B | 17.9 |
| 6 | Qwen 3 235B A22B | 16.7 |
| 7 | DeepSeek V3.2 | 15.6 |
| 8 | Llama 3.1 8B | 15.6 |
| 9 | Qwen 2.5 72B | 14.7 |
Interactive version: theaggregate.ai/benchmark?slug=stresseval-text-knowledge-stress · How It Works · Data refreshed daily, snapshot 2026-10-07.