CLBench Life - Communication & Social Interactions: leaderboard
CL-bench Life category 'Communication & Social Interactions': context-learning tasks grounded in communication and social-interaction records from everyday life. Solving rate in percent (published as a 0-1 fraction), higher is better; labels keep the reasoning effort the source records.
Metric: Solving Rate (%). Source: www.clbench.com. Saturation forecast: Around 2029. 29 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.5 (High) | 23.5 | #33 (GPT-5.5) |
| 2 | GPT-5.4 (xHigh) | 23 | #83 (GPT-5.4) |
| 3 | GPT-5.4 (High) | 22.2 | #83 (GPT-5.4) |
| 4 | Claude Opus 4.6 (High) | 21.7 | #67 (Claude Opus 4.6) |
| 5 | Gemini 3.1 Pro (Preview) | 21.5 | #64 |
| 6 | GPT-5.1 (High) | 21.5 | #145 (GPT-5.1) |
| 7 | Claude Opus 4.6 | 20.7 | #67 |
| 8 | Seed 2.0 Pro (High) | 19.8 | #122 (Seed 2.0 Pro) |
| 9 | Gemini 3.1 Pro (Preview) (High) | 18.8 | #64 (Gemini 3.1 Pro (Preview)) |
| 10 | GPT-5.4 | 17 | #83 |
| 11 | Hy3-preview (High) | 17 | #166 (Hy3-preview) |
| 12 | Kimi K2.5 (High) | 15.3 | #152 (Kimi K2.5) |
| 13 | Hy3-preview (Non-reasoning) | 13.3 | #166 (Hy3-preview) |
| 14 | Grok 4.20 (High) | 12.8 | #179 (Grok 4.20) |
| 15 | DeepSeek V4 Pro (High) | 12.6 | #112 (DeepSeek V4 Pro) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=clbench-life-communication-social-interactions · How It Works · Data refreshed daily, snapshot 2026-11-05.