customnodes 开源项目推荐（14 个）

1038lab /ComfyUI-QwenVL

ComfyUI-QwenBL自定义节点：集成了Qwen-BL系列（包括Qwen 2.5-BL和最新的Qwen 3-BL），并支持GGUF在文本生成、图像理解和视频分析方面的高级多模式人工智能。

1038lab /ComfyUI-FlashVSR

Powerful ComfyUI custom node built on the FlashVSR V1.1 model, facilitating real-time diffusion-based video super-resolution for streaming applications.

1038lab /ComfyUI-MiniMax-Remover

ComfyUI-MiniMax-Remover is a custom node for ComfyUI that enables fast and efficient object removal using minimax optimization. It works in two stages: first, it trains a remover with a simplified DiT model; then it distills a robust version using CFG guidance and fewer inference steps.

1038lab /ComfyUI-VoxCPMTTS

一个干净高效的ComfyUI定制节点，用于VoxCPM TTS（文本转语音）功能。该实现采用VoxCPM 1.5模型，提供高质量的语音生成和语音克隆能力。

1038lab /ComfyUI-FireRedTTS

A ComfyUI integration for FireRedTTS‑2, a real-time multi-speaker TTS system enabling high-quality, emotionally expressive dialogue and monologue synthesis. Leveraging a streaming architecture and context-aware prosody modeling, it supports natural speaker turns and stable long-form generation, ideal for interactive chat and podcast applications.

1038lab /ComfyUI-DeepSeek-OCR

A powerful OCR node for ComfyUI that integrates the DeepSeek-OCR model from Hugging Face.

KoreTeknology /ComfyUI-Nai-Production-Nodes-Pack

A set of Custom Nodes for Compositing for ComfyUI

RiverSide71 /comfyui-library-finder

A custom node which finds which custom nodes are importing a particular library.

1038lab /ComfyUI-API-Manager

LLM API Manager for ComfyUI

inkbottle-9 /comfyui_dynamic

Adds a full-featured Python script execution node and some quality-of-life nodes to ComfyUI

MazenZya /playport

Run PC games natively on your iPhone with Playport — no streaming. Open-source iOS app with game compatibility tracking and installable releases.

Paata28b /Qwen3.5-9B-ToolHub

Deploy a local Qwen3.5-9B multimodal AI with GPU inference supporting web search, image queries, file reading, and an OpenAI-compatible API.

labestia2 /playport

Run wireless CarPlay on a Mac and stream it to any browser on your network—touch, media keys, Siri, and calls, no dongle needed.

lergah /qwen36-27b-single-3090

Run Qwen3.6-27B models on a single RTX 3090 GPU using optimized inference configs. See club-3090 for active developments and updated configurations.