OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Frameworks

Comfyui LLM Party

Comfyui LLM Party is an open-source collection of Python nodes for building language-model and agent workflows in ComfyUI and connecting them with image workflows. It includes documented paths for API and local language or vision-language models, plus integrations such as MCP and GraphRAG.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Comfyui LLM Party — OpenViST project profile in Agent Frameworks.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when ComfyUI is already part of your workflow and you want to add model calls, agent steps or retrieval alongside image processing. Start with a supplied workflow that matches your API, Ollama or local-model setup.

Who it is for

  • ComfyUI users adding LLM calls or agent steps to image workflows.
  • People configuring workflows with API-hosted models or local models through Ollama, distributed formats or GGUF.
  • Workflow builders exploring retrieval, tool use, OCR, text-to-speech or multi-agent patterns inside ComfyUI.

Documented uses

  • Call hosted or local language models from a ComfyUI workflow.
  • Use a language or vision-language model to help generate prompts and images.
  • Connect model workflows with tools, MCP integrations or retrieval over a knowledge base.
  • Combine LLM stages with image workflows and documented OCR or text-to-speech nodes.
Before you choose

Requirements & limits

The project is a public GitHub repository with an AGPL-3.0 license. The README describes a Windows-only portable package as one installation option; it also explains adding the nodes to an existing ComfyUI installation.

  • ComfyUI is the workflow front end, so familiarity with its workflow and node model is relevant.
  • API model configuration requires a base URL and API key; local model routes require the corresponding local setup and model details.
  • The README directs users to install missing nodes through ComfyUI-Manager and notes that dependency issues can arise.
  • The project supplies separate starter workflows for API calls, Ollama, distributed local models, GGUF and local vision-language models.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up a model-call workflow in ComfyUI

Load a documented starter workflow, install its missing nodes and configure an API or Ollama model connection.

Before you start

  • An existing ComfyUI installation; alternatively, the README describes a Windows-only portable package that includes ComfyUI and LLM Party.
  • ComfyUI-Manager to install missing nodes.
  • For API calls, a base URL and API key. For Ollama, an Ollama model and its local endpoint.

Steps

  1. Drag a supplied starter workflow, such as the API-call or Ollama workflow, into ComfyUI.

  2. Use ComfyUI-Manager to install any missing nodes.

  3. For an API model, enter the base URL and API key in the API LLM loader node. For Ollama, enable its Ollama option; alternatively, configure the documented local endpoint and model name.

Expected outcome

A ComfyUI workflow is configured to call an API-hosted or Ollama-managed language model.

Guide limitations

  • This guide covers setup and configuration only; the README does not provide a verified execution result.
  • The portable package described in the README is limited to Windows.
  • The API route requires credentials and an endpoint; the Ollama route requires a local Ollama setup.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
9 October 2026
Original website title
heshengtao/comfyui_LLM_party · Official homepage title · retained in its source language
How we review sourcesSuggest a correction