CLBench Life - Personal Information Fragments: leaderboard

CL-bench Life (2026) tests whether language models can learn from long, messy real-life context to solve human-written tasks, a task counting as solved only when every expert rubric item is met. This board is the task solving rate in percent on tasks over personal notes, diaries, to-do lists and saved messages; higher is better.

Metric: Task solving rate (%) on the 45 personal-information-fragment (notes, diaries, to-do lists, saved messages) tasks of CL-bench Life; a task is solved only when the response satisfies every human-written binary rubric item, judged by GPT-5.1 at high reasoning effort from the response and rubrics alone; reasoning mode at high effort where the model supports it, mean of three runs; higher is better. Source: arxiv.org. Saturation forecast: Around 2030. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Hy3-previewHigh20#322
2GPT-5.4High14.8#135
3Claude Opus 4.6High14.8#113
4Grok 4.20High13.3#301
5Kimi K2.5High11.1#266
6Qwen 3.5 PlusHigh11.1#242
7Gemini 3.1 Pro (Preview)High10.4#99
8Seed 2.0 ProHigh10.4#222
9DeepSeek V3.2Thinking8.1#419
10MiniMax-M2.5Thinking6.7#600

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

Interactive version: theaggregate.ai/benchmark?slug=clbench-life-personal-information-fragments · How It Works · Data refreshed daily, snapshot 2026-11-21.