Millions of tasks, ten thousand concurrent, GoPool outperforms the ten-thousand-star GitHub project ants and the thousand-star project pond.
| Project | Time to Process 1M Tasks (s) | Memory Consumption (MB) |
|---|---|---|
| GoPool | 1.13 | 1.88 |
| ants(10k star) | 1.43 | 9.49 |
| pond(1k star) | 3.51 | 1.23 |
Everyone is welcome to learn more about the GoPool project and try GoPool. Feedback through issues and contributions via PRs are also greatly appreciated.
Hi @daniel-hutao,
Sounds interesting! Do you have a link to the repo?
And how did you utilize GPT to develop the project?
prompts can be found here, but they are Chinese now. I’ll translate them to English next week.
I can not read Chinese but I can tell that ChatGPT did not “write” this. It filled in some blanks in code you wrote. Your claims misrepresent what happened. The prompts are basically CODE templates and pseudo code. ChatGPT can not “write code” it is just doing a MadLibs and that has been done for years without LLMs.
I don’t understand what you are saying. First of all, I am not using ChatGPT, but gpt-4; secondly, ChatGPT can write code; then I did not use gpt-4 directly, but used DevChat, which will add some extra messages before sending messages to gpt-4. I have published the complete prompts, but in Chinese. I will translate it into English soon.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.