EvoClawBench - Baseline (nanobot): leaderboard
Metric: Execution-only mean task score (%; 100 fixture-backed tasks plus a sanity task, 502 sub-problems across coding, data, office, security, operations and document workflows, automated or hybrid graders; one run per task in the nanobot agent runtime, execution phase only; the agent solves each task directly and may not create skills). Source: arxiv.org. Saturation forecast: Rough model projection: around 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | nanobot + GPT-5.4 | 96.13 |
| 2 | nanobot + MiniMax-M2.7 | 90.97 |
| 3 | nanobot + GPT-5.4 Mini | 88.71 |
| 4 | nanobot + DeepSeek V4 Pro | 77.77 |
| 5 | nanobot + Qwen 3.6 Plus | 56.45 |
Interactive version: theaggregate.ai/benchmark?slug=evoclawbench-baseline-nanobot · How It Works · Data refreshed daily, snapshot 2026-09-29.