LiveClawBench - Easy: leaderboard

Agent performance on the 69 Easy cases of LiveClawBench (2026), personal-assistant agent tasks in reproducible full-stack mock services; a case is Easy when the anchor model DeepSeek-V4-Pro earns a mean reward above 0.7 on it. Average reward over three runs, from 0 to 100; higher is better.

Metric: Average reward (0-100) over three runs on the 69 Easy cases (DeepSeek-V4-Pro anchor reward above 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: Estimated already saturated. 16 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Kimi K2.7 Code95.6#208
2Qwen 3.6 Plus95.3#245
3GLM-5.195.3#200
4MiniMax-M2.794.8#539
5Qwen 3.6 Flash94.3#454
6GPT-5.593.4#56
7Claude Opus 4.893.1#64
8MiniMax-M392.3#290
9Kimi K2.691.8#197
10GLM-5.291.6#154
11MiMo-V2.5-Pro90.7#292
12Qwen 3.6 27B82.3#358
13Qwen 3.5 Plus82#250
14DeepSeek V4 Flash81.9#273
15Qwen 3.5 Flash79.5#447

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=liveclawbench-easy · How It Works · Data refreshed daily, snapshot 2026-11-22.