LiveClawBench - Medium: leaderboard
Metric: Average reward (0-100) over three runs on the 46 Medium cases (DeepSeek-V4-Pro anchor reward 0.3 to 0.7), OpenClaw-style personal-assistant tasks run in reproducible full-stack mock services (22 stateful web applications, browser, APIs and shell), three runs per case, rewards from case verifiers and a DeepSeek-V3.2 rubric judge for open-ended cases; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 17 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 67.7 |
| 2 | Kimi K2.7 Code | 67.3 |
| 3 | GLM-5.2 | 65 |
| 4 | GLM-5.1 | 63.2 |
| 5 | MiMo-V2.5-Pro | 62.6 |
| 6 | MiniMax-M3 | 59 |
| 7 | Qwen 3.6 Plus | 57.3 |
| 8 | Qwen 3.6 Flash | 57.2 |
| 9 | Claude Opus 4.8 | 56.6 |
| 10 | Kimi K2.6 | 55 |
| 11 | MiniMax-M2.7 | 48.9 |
| 12 | DeepSeek V4 Flash | 48.1 |
| 13 | Qwen 3.5 Plus | 37.6 |
| 14 | Qwen 3.5 Flash | 34.6 |
| 15 | Qwen 3.6 27B | 33.4 |
Interactive version: theaggregate.ai/benchmark?slug=liveclawbench-medium · How It Works · Data refreshed daily, snapshot 2026-10-07.