lagent issues

1

I really like your framework and currently trying locally and evaluate now. Does Lagent framework support Ollama?If not , is it possible to work around and use my local Ollama...

yzhao244

Question for React Agent

1

I have taken a look at inside lagent_example.py and react.py. it seems like the react prompt does not incorporate the previous output as new input. Is it a systematic mistake,...

aesashi

【feature】llm添加openai风格服务端支持(自定义代理地址和模型名称)

1

# llm添加openai风格服务端支持(自定义代理地址和模型名称) 使用mindsearch的时候，发现无法使用自定义代理地址、模型名称的openai风格服务端。 GPTAPI有进行模型校验等原因，秉承只add的原则。基于GPTAPI扩展GPTStyleAPI： ## 支持openai风格：已测试通过： - xinference 1.2.0 - ollama 0.5.5 - one-api v0.6.10-alpha.6 - baichuan 直连 - lmdeploy 0.7.0 ## 使用方法：设置api_base、 model_name、key（如果服务端不需要，可忽略） ``` api_base = 'http://192.168.26.213:13000/v1/chat/completions'...

QiuZiXian

API is valid but got invalid response, I am not sure where I got it wrong.

9

API is valid but got invalid response, I am not sure where I got it wrong. ``` import os import logging import warnings as wa wa.warn_explicit = wa.warn = lambda...

ghLcd9dG

openai 中的 GPTAPI 的参数配置错误导致除非流式外的智能体无法处理通义千问模型的响应

1

# 所用版本提交 #285，main 分支。 # 问题概述使用如下代码尝试获取智能体回复时： ```python from lagent.agents import Agent from lagent.schema import AgentMessage from lagent.llms import GPTAPI # qwen 智能体配置来自项目 MindSearch url = "https://dashscope.aliyuncs.com/api/v1/services/aigc/text-generation/generation" qwen =...

wojiaoyishang