Deepseek official api, change the id to deepseek-v4.1-flash-expires-on-0910 and it can be used, the speed is really fast
Deepseek-v4.1-flash first released for Pelican testing!
Thanks to the average speed of more than 300 tokens/s, this test was completed very quickly. The most time-consuming thing was that my typing and screen recording were too slow! Sure enough, humans are the biggest sticking point for AI, hahahahahaha If you exclude the special training for this topic, I think the effect of this implementation is still good.
Deepseek-v4.1-flash-expires-on-0910's Pelican Ride, generated almost instantly
I dug out the AI shell that I used for myself. At that time
I dug out the AI shell that I used for myself. At that time, the concept of MCP was just becoming popular (it was still the era of DeepSeek V3 and R1). In order to use it smoothly and learn related concepts, I spent a month working on it. It does not use any off-the-shelf AI SDK or LangChain.
You have reached the end of the archive
All of deepseek