-
部署方式:使用源码构建docker镜像
在使用通义的api接口时就一直提示fetch error,控制台提示跨域。做了多种尝试都没有解决,目前推测因为通义的API比openai的多了一级目录 https://dashscope.aliyuncs.com/**compatible-mode** 导致的,kimi这种完全跟openai层级一样的就没问题。可能是做跨域或代理的时候没有适配阿里的地址…
-
### 🥰 需求描述
目前的通义千问似乎使用的是openai兼容接口,导致除qwen系列模型外,阿里云的其他模型无法兼容,能否支持更多模型的使用?
![图片](https://github.com/user-attachments/assets/c37020c8-d26b-4e31-a92f-432846c5c9de)
### 🧐 解决方案
改用dashscope的接口应该可以解决这个…
-
not work with `qwen-vl-max` as provide not supported param: `temperature`
code:
```py
from GeneralAgent import Agent
models = [
('qwen-vl-max', 6000, 'sk-0xxx', 'https://dashscope.aliyu…
-
**问题描述 / Problem Description**
需要以非openAI格式调用其他的llm api,我自己配置了一个worker,但对话时一直报错
**复现问题的步骤 / Steps to Reproduce**
执行LLM chat 报错 utils.py[line:41] - ERROR: TypeError: Caught exception: object of ty…
-
我内网环境部署的qwen模型,需要设置一下extra_generate_cfg = {
"extra_headers": {
"Host": "qwen1-5-32b-isvc.ai-platform.example.com"
}
}
来指定Host,我如果使用FnCallAgent类来实现智能体,配置好llm_cfg 参数…
-
**AgentScope is an open-source project. To involve a broader community, we recommend asking your questions in English.**
**Describe the bug**
A clear and concise description of what the bug is.
…
-
simply review a few sections of our code and write the readme about it.
-
今天通义千问大降价,似乎可以自己买个Qwen-Long API key试试了。
-
app.py:
![image](https://github.com/user-attachments/assets/5d4f47c6-eafe-4b0d-9916-8318aa500efa)
streamlit:
![image](https://github.com/user-attachments/assets/4f4ce77d-9765-47c0-bb88-5ce17623…
-
### Problem Description
I don't know how to use the Qianwen model's API endpoint. Does this software not support the Qwen model?
### Solution Description
## Please let me know if it is my invocatio…