Multi agent system for AI-driven software development. Combine LLM with DevOps tools to convert natural language requirements into working software. Supports any development language and extends the existing code.
I replaced the OpenAI code in "llm_basic.py" with a request from the RWKV-Runner API and tested using model "RWKV-4-World-3B-v1-20230619-ctx4096" but I got some errors after running a simple task, although the API configured correctly, I tested it using a prompt from "subtask_basic.py" and the returned ["choices"][0]["message"]["content"] from results looks correct like the "rwkv_response.json" window in screenshot
We use prompt to control the response format of the language model. Different language models do not understand the same prompt in the same way. In the open source version we recommend using gpt3.5
I replaced the OpenAI code in "llm_basic.py" with a request from the RWKV-Runner API and tested using model "RWKV-4-World-3B-v1-20230619-ctx4096" but I got some errors after running a simple task, although the API configured correctly, I tested it using a prompt from "subtask_basic.py" and the returned ["choices"][0]["message"]["content"] from results looks correct like the "rwkv_response.json" window in screenshot