# comfyui_LLM_party

> MCP Client

Comfyui_llm_party provides ComfyUI nodes for building LLM workflows, enabling users to quickly create and integrate them into image workflows. It offers a complete set of tools for LLM workflow construction.

## Overview

- **Category:** AI
- **Language:** Python
- **Stars:** 2311
- **Forks:** 133
- **Owner:** heshengtao
- **GitHub:** https://github.com/heshengtao/comfyui_LLM_party
- **Created:** 2024-04-13T13:42:49+00:00
- **Updated:** 2025-03-28T16:37:14+00:00
- **Source:** https://model-context-protocol.com/clients/comfyui-llm-agent-workflow-image-generation

## Setup

## Setup
1. If you have never used ComfyUI and encounter some dependency issues while installing the LLM party in ComfyUI, please click [here](https://drive.google.com/file/d/1T9C7gEbd-w_zf9GqZO1VeI3z8ek8clpX/view?usp=sharing) to download the **Windows** portable package that includes the LLM party.
2. Drag the workflows into your comfyui, then use [comfyui-Manager](https://github.com/ltdrdata/ComfyUI-Manager) to install the missing nodes.
3. If you are using API, fill in your `base_url` and `api_key` in the API LLM loader node.
4. If you are using ollama, turn on the `is_ollama` option in the API LLM loader node, no need to fill in `base_url` and `api_key`.
5. If you are using a local model, fill in your model path in the local model loader node.
6. Navigate to the `comfyui_LLM_party` project folder.
7. Enter `pip install -r requirements.txt` in the terminal to deploy the third-party libraries required by the project into the comfyui environment.
8. If you are using the comfyui launcher, you need to enter `path_in_launcher_configuration\python_embeded\python.exe -m pip install -r requirements.txt` in the terminal to install.
9. The language can be configured in `config.ini`.
10. APIKEY can be configured in `config.ini` or directly in the LLM node.

## Tools

## Available Tools

1.  LLM multi-tool call (basic functionality).
2.  Role setting (quickly build your own exclusive AI assistant).
3.  Industry-specific word vector RAG and GraphRAG (localize the management of the industry knowledge base).
4.  Single agent pipeline (basic agent setup).
5.  Complex agent-agent radial interaction mode and ring interaction mode (advanced agent interaction).
6.  Access to social APP (QQ, Feishu, Discord) (connect to social apps).
7.  One-stop LLM + TTS + ComfyUI workflow (for streaming media workers).
8.  Parameter debugging interfaces (for scientific researchers, model adaptation).
9.  Streaming output mode (display the text returned by the API in real-time).
10. Reasoning_content output (automatically separate the reasoning and response of the R1 model).
11. Image hosting node (supports image hosting services at https://sm.ms and https://imgbb.com).
12. MCP tool (connect to the MCP server and convert the tools from the server into tools that can be directly used by LLMs).