Skip to content

Make images and video without living inside the AI toolchain.

A local-first workspace for turning ComfyUI workflows into tools you can use, understand, and reuse.

Product

Download Studio Resources Workflow Doctor Agent Access

Learn

Docs FAQ Updates

Connect

Feedback Privacy Terms

© 2026 Tea Time Labs

Local AI creation, organized.

Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗

Download StudioLanguage

SCG_LocalVLM | ST Studio

Resources/SCG_LocalVLM

Node packListedExternal

SCG_LocalVLM

ComfyUI QwenVL and Qwen wrapper

ComfyUI QwenVL and Qwen wrapper

Sourcegithub.com/SanDiegoDude/SCG_LocalVLM

Maintainer@SanDiegoDudeCgrade

Package profile

Status: Listed
Publisher: @SanDiegoDude
Source: GitHub
Stars: 6

README

ComfyUI QwenVL and Qwen wrapper

Nodes in this pack

4 nodes

Qwen

Sources

1 source

GitHubSource

github.com/SanDiegoDude/SCG_LocalVLM

Verified Sep 3

Source excerpts

1 excerpt

NoteJul 16github.com/SanDiegoDude/SCG_LocalVLM

Source context: Repo SanDiegoDude/SCG_LocalVLM

Qwen Text-Only Language Model Node for ComfyUI Supports Qwen2.5 and Qwen3 text generation models with: - Multiple quantization options (none, 4bit, 8bit) - Advanced generation parameters - Flash Attention 2 optimization - System and user prompt support ComfyUI v2/v3 compatible.

Comfyui QwenVL · 1 output

Qwen2.5

Qwen2.5

Compact description unavailable for this node.

Qwen2.5VL

Qwen2.5VL

Compact description unavailable for this node.

QwenVL

QwenVL

Qwen Vision-Language Model Node for ComfyUI Supports Qwen2.5-VL and Qwen3-VL models with: - Multi-image input (up to 4 images) - Video input via FFmpeg processing - Multiple quantization options (none, 4bit, 8bit) - Advanced generation parameters - Flash Attention 2 optimization ComfyUI v2/v3 compatible.

Comfyui QwenVL · 1 output