# Rohan Paul 评 Atomic Bot 实验：OpenClaw 2.0 与 Hermes Agent 在 GLM 5.3 上的差异主要在自检方式

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-09-01 10:58
- AIHOT 分数：49
- AIHOT 链接：https://aihot.virxact.com/items/cmti3kurp098zrofqql65qi2k
- 原文链接：https://x.com/rohanpaul_ai/status/2094620857981399131

## AI 摘要

Rohan Paul 转引 Atomic Bot 的对比实验：在 GLM 5.3 上用四个提示词让 OpenClaw 2.0 与 Hermes Agent 生成以 OpenClaw 吉祥物为主角的电影场景，OpenClaw 2.0 消耗约 2.1M tokens、约 $4.5、10 次自修复，Hermes Agent 消耗约 2.9M tokens、约 $4、20 次自修复。

## 正文

OpenClaw 2.0 vs Hermes experiment by @atomicbot_ai is a good example of why the model alone tells you very little about agent performance.

Outputs:
OpenClaw 2.0: ~2.1M tokens, ~$4.5, 10 self-fixes
Hermes Agent: ~2.9M tokens, ~$4, 20 self-fixes

The difference was in how they checked themselves.

Atomic Bot lets you run AI agents in the cloud without managing the setup yourself.

In this experiment

• OpenClaw screenshotted frames at specific timestamps, counted pixels, then corrected visual errors such as an inverted coat-tail direction.
• Hermes went further and created its own render-checking setup, measuring frame timing, testing whether loop seams were visible, and auditing scenes against the original spec.

That extra ~800K tokens was largely verification work.

We usually talk about agent tokens as inference cost.

This run is a nice reminder that some of those tokens are effectively a QA budget. Once agents can generate their own checks, the important question becomes how much compute you want them spending proving their work before returning it.

### 引用推文

> atomicbot.ai：OpenClaw 2.0 is available on Atomic Bot! We ran a battle: OpenClaw 2.0 vs Hermes Agent on GLM 5.3. Four prompts: movie scenes with the OpenClaw mascot in the le...