LLMApi Cli

Talk to LLMs like ChatGPT in command line

Introduction

中文文档

llmapi_cli is a command-line tool that can directly chat with chatgpt/gpt3 and more large language models(LLM), based on the llmapi-server interface. At the same time llmapi_cli also provides Python module (LLMClient).

For more information visit: llmapi.io

Install

# install from pypi
python3 -m pip install llmapi_cli

# install locally
python3 setup.py install

Usage

# The first time you use this command,
# you need to specify parameters, this 
# will be cached after successful connection.
#
# If you use your own llmapi-server:
llmapi_cli --host='http://127.0.0.1:5050' --bot=mock
#
# If you use api.llmapi.io (which is default host):
llmapi_cli --bot=mock --apikey='your apikey on llmapi.io'

# You can directly use this command again:
llmapi_cli
#
# Or change the bot type:
llmapi_cli --bot=gpt3
#
# Or redesignate params like first time.

Screenshot

Currently supported bot type

View llmapi-server for more information.

Warning: Currently for testing only

• chatgpt: openai's official ChatGPT,see more
• gpt3: openai's official GPT-3
• welm: Tencent weixin's llm,see more
• newbing: microsoft's new bing chat, UNOFFICIAL INTERFACE