LongNovel (32K): leaderboard

Metric: Balanced accuracy (%; M, about 32k tokens context scale; binary detection of hallucinated summaries of Chinese novels and English BookSum chapters (8 hallucination types, human-revised test set); direct prompt with the summary at the end; an output without a valid consistency score counts as an incorrect prediction; chance is 50). Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.

Top models

#ModelScore
1DeepSeek V4 Flash81
2GPT-5.2 Instant77
3Claude Sonnet 4.566.5
4Qwen 3 8B60.8
5Qwen 3 14B59.8
6Qwen 3 32B57.5
7GLM-4 9B Chat51
8internlm2.5-20B-chat50.5

Interactive version: theaggregate.ai/benchmark?slug=longnovel-32k · How It Works · Data refreshed daily, snapshot 2026-09-29.