StressEval - Knowledge Graph Knowledge Stress: leaderboard

StressEval (2026) builds Dynamic-OneEval by synthesizing new knowledge-intensive questions from GPT-5.2 failure cases on HotpotQA, ComplexWebQuestions and Wikipedia table question sets, keeping the difficulty factor that caused each failure. This board is exact-match accuracy in percent on knowledge-graph questions that need a fact withheld from the graph; higher is better.

Metric: Exact-match accuracy (%) on the knowledge-graph knowledge-stress split (432 instances seeded from ComplexWebQuestions, answering needs a fact withheld from the graph) 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: Not forecast. 9 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 2.5 72B33.9#1188
2Gemini 3 Pro28.6#145
3Llama 3.1 70B27.8#1499
4QwQ-32B22.4#907
5DeepSeek V3.215.7#419
6Claude Sonnet 4.513.9#261
7Qwen 3 235B A22B13#628
8Llama 3.1 8B11.2#5196
9Seed-1.61.9#469

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=stresseval-knowledge-graph-knowledge-stress · How It Works · Data refreshed daily, snapshot 2026-11-21.