DeepSeek V4 Flash needs 2× RTX 6000s
Qwen3.8-27B FP8 needs just 2× RTX 4090s (2,873 tok/s prefill, 74 tok/s decode) I was curious how they compared, so I ran the same Pi agentic task on both. Both solved it in one shot with no follow-ups Here are the outputs side by side👇
What makes me feel very interesting is the “track” behind DeepSeek Harness
You can clearly see the thinking behind the model How raw input and output becomes user UI step by step And you can view the conversation details and Token consumption in chronological order This reminds me of Chrome’s Dev tool The idea that everything is a plug-in also reminds me of Chrome…
Want to add functionality to the DeepSeek agent, but don’t know where to find it?
This directory site helps you find them all in one go! 👍🔥 Within a few days of the release of DeepSeek Harness, plug-ins exploded, but the resources were too scattered and difficult to find. 😭 Someone has compiled a selection directory site, which currently contains 160 plug-ins, divided into 11 categories (interface, memory, tools, skills, workflow, etc.). Each plugin has an installation command and GitHub…
Wow! All we can say is that as soon as DSH came out, heroes from all walks of life rushed to appear.
cgnot996, the author of Grok App, DeepSeek APP is also released! And integrates various capabilities of Grok, Let X become DeepSeek's knowledge base. Is this the highest mountain in the world? roar! ha!
You have reached the end of the archive
All of deepseek