DeepSeek Super Agent vs Manus AI
Comparer des modèles
DeepSeek Super Agent is more worth it — 60% vs 50%
How we score worth-it? →DeepSeek's multi-task super agent that splits and handles multiple tasks at once.
- My current setup: deepseek superagent harness.— hn ↗
- On a real project, dsh got every feature done! UI restoration, architecture design, frontend code, backend code, complex atomic tasks, committing code to GitHub, logging into the ECS server to deploy, troubleshooting production issues, checking server logs, fixing and pushing code, deploying!— v2ex ↗
- Except that DeepSeek Harness is very well done. It is not just the model.— hn ↗
- Exactly — on the vast majority of tasks, ds+dsh is already no different from the frontier, and the experience is even better in other dimensions: faster, more agile, no various reviews.— v2ex ↗
- Same text in, text out — dsh shell basically gives you natural-language input and output, and the UI and extensibility are pretty good too.— v2ex ↗
- The GitHub repository itself stresses this is a research prototype, not a production-ready product, and may contain bugs.— github ↗
- I compared them: for the same task with the same model (mostly ds and gpt), dsh uses almost twice the tokens and time to finish, and the delivered result isn't better than pi's.— v2ex ↗
- I just don't install plugins at all, everything goes through MCP externally; dsh has too many breaking upgrades and plugins break constantly.— v2ex ↗
- dsh itself is a rough scaffold right now, and the plugin ecosystem is full of junk.— v2ex ↗
“Multi-task super agent that splits and handles several tasks at once”
“Authorization flaw in the management API allows local code execution (High severity)”
General-purpose autonomous agent
- I like it and use it regularly. Maybe I don't get other harnesses, but in my experience Manus gets shit done, whereas I have to constantly babysit other tools.— hn ↗
- Manus 2.0 is good at many long time run tasks. I prefer one agent researched, the other narrowed the list.— youtube ↗
- There is something still very unique with Manus and that is why Meta wanted to do business. I enjoy Manus and Cue.— youtube ↗
- 我的实习生只和 Manus 聊了三轮,做出两条模拟恐怖视频,抖音播放 800 万,新号 7 天涨粉 1.6 万。— bilibili ↗
- Manus 限时免费,我就用它做了 2 个论坛脚本, 1 个邮件助手插件,方便个人使用!— v2ex ↗
- But Manus's hype is quite repulsive.— v2ex ↗
- I used 30% of my monthly credits in ONE session.— blog ↗
- My one-year Manus Pro subscription hadn't even expired yet and they just cancelled it — zero business ethics.— v2ex ↗
- The search quality is too low — it's all Sina and Tencent News items pulled from Baidu. That material is low value, and it feels nowhere near usable for real production work.— bilibili ↗
- I hope it has better quality than ManusAI— hn ↗
“The demos are unreal — multi-step browser tasks done end-to-end.”
“In practice it's more hit-or-miss. Some tasks it crushes, some it fumbles.”
“Credit-based pricing means you're always watching the meter.”
Which should you pick?
The gap is clear: DeepSeek Super Agent sits at 60% while Manus AI is at 50%, a 10-point difference. If you just want the safer pick, start with DeepSeek Super Agent.
They price differently (DeepSeek Super Agent: Free (MIT); Manus AI: From $20/mo (credit)), so estimate by your own usage.
Both are in the Agent category with similar coverage; the real difference is stability and cost — run one of your own tasks through each before deciding.
What users actually said about DeepSeek Super Agent: "My current setup: deepseek superagent harness."
FAQ
Is DeepSeek Super Agent or Manus AI more worth it?
DeepSeek Super Agent is more worth it, with a 60% worth-it score vs 50% for Manus AI.
How much does DeepSeek Super Agent cost vs Manus AI?
DeepSeek Super Agent: Free (MIT) · Manus AI: From $20/mo (credit).
What do users say about DeepSeek Super Agent vs Manus AI?
DeepSeek Super Agent: Multi-task super agent that splits and handles several tasks at once. Manus AI: The demos are unreal — multi-step browser tasks done end-to-end..