supermarket

Benchmark

CL-bench Life

Measures: A companion to CL-bench with messy real-life context (everyday messages, fragmented notes, activity trails): 405 tasks the model completes by learning from that context.

Maintainer
Tencent Hunyuan and Fudan University NLP group
Introduced
2026
Status
Active
Saturation
Headroom
Typical provenance
Independent
In scoring set
No

PaperOfficial leaderboard

Leaderboard

Show independent results only

Every result we hold, whoever ran it. Only independent and third-party rows count towards a score.

Leaderboard
ModelTasks solvedProvenanceSourceEvaluated
GPT-5.5
OpenAI · High effort
22.2%Source not statedEpoch AI via Epoch AI—
GPT-5.4
OpenAI · Xhigh effort
21.7%Source not statedEpoch AI via Epoch AI—
GPT-5.4
OpenAI · High effort
19.3%Source not statedEpoch AI via Epoch AI—
GPT-5.1
OpenAI · High effort
17.3%Source not statedEpoch AI via Epoch AI—
Claude Opus 4.6
Anthropic · High effort
17.0%Source not statedEpoch AI via Epoch AI—
Gemini 3.1 Pro
Google DeepMind
16.9%Source not statedEpoch AI via Epoch AI—
GPT-5.4
OpenAI · Unknown effort
13.8%Source not statedEpoch AI via Epoch AI—
Claude Opus 4.6
Anthropic · Unknown effort
13.6%Source not statedEpoch AI via Epoch AI—
DeepSeek-V4-Pro
DeepSeek · High effort
13.5%Source not statedEpoch AI via Epoch AI—
Kimi K2.5
Moonshot AI
13.2%Source not statedEpoch AI via Epoch AI—
Qwen3.5-Plus
Alibaba Group
12.4%Source not statedEpoch AI via Epoch AI—
Grok 4.20
SpaceXAI (xAI)
11.9%Source not statedEpoch AI via Epoch AI—
GLM-4.7
Z.ai (Zhipu AI)
10.9%Source not statedEpoch AI via Epoch AI—
DeepSeek-V3.2-Exp
DeepSeek · Thinking effort
9.5%Source not statedEpoch AI via Epoch AI—
DeepSeek-V3.2
DeepSeek
7.4%Source not statedEpoch AI via Epoch AI—
MiMo-V2-Pro
Xiaomi
6.9%Source not statedEpoch AI via Epoch AI—
MiniMax-M2.5
MiniMax
6.3%Source not statedEpoch AI via Epoch AI—
  • Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.

Sources