Introduction

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

Config for your environment

Replace {MCP_ENDPOINT_URL} with this MCP’s endpoint URL (from its repo or docs above). No API key — you connect directly.

Tool

Config file: ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp-server": {
      "url": "{MCP_ENDPOINT_URL}"
    }
  }
}

Paste into mcpServers in the config file. Restart Cursor after saving.

llmapi-cli

Links

README