Getting Started with Qwen Lightning v1.1 in ComfyUI for 8GB VRAM
Jean

Jean @aitechtutorials

About: Teaching how to use AI tools like Gemini CLI and MCP through step-by-step YouTube tutorials. Building, breaking, and simplifying AI workflows.

Joined:
Jul 1, 2025

Getting Started with Qwen Lightning v1.1 in ComfyUI for 8GB VRAM

Publish Date: Aug 15
0 0

Artificial intelligence image generation has never been easier or faster than with Qwen Lightning v1.1 in ComfyUI. Whether you’re a beginner or an experienced AI artist, this guide walks you through every step of installation, configuration, and optimization, ensuring a smooth workflow and stunning results.

🎥 Watch the full step-by-step video tutorial here: https://youtu.be/7p23f-gDp4I

Step 1: Download the Necessary Models

GGUF Model: Download the Qwen Lightning v1.1 GGUF model for 8GB VRAM.

Text Encoder: Download the FP8 scaled text encoder.

VAE: Download the associated VAE file.

These components are essential for generating high-quality, detailed AI images.

Step 2: Organize Your Models

Place the downloaded files in the appropriate ComfyUI folders:

GGUF → ComfyUI/models/unet

Text Encoder → ComfyUI/models/text_encoders

VAE → ComfyUI/models/vae

LoRA → ComfyUI/models/loras

Correct placement ensures ComfyUI can access all resources seamlessly.

Step 3: Update ComfyUI and Load the Workflow

Open the ComfyUI Manager and click Update All to ensure the UI, nodes, and templates are up-to-date.

Download the example workflow JSON from Hugging Face.

In ComfyUI, navigate to Workflow → Open and select the JSON file.

Assign loaders: UNet = GGUF, VAE = VAE, CLIP = FP8 scaled.

Step 4: Install and Configure Qwen Lightning v1.1 LoRA

Place the LoRA file in the loras folder.

Load it using the LoRA Loader node.

Adjust the strength (0.7–0.9 to start) to influence the image generation while maintaining clarity and style.

Step 5: Optimize for Speed and Quality

Enable Sage Attention and FP16 Accumulation to reduce VRAM usage and speed up generation.

Adjust CFG between 4–4.5 to balance quality and rendering speed.

With these settings, Qwen Lightning v1.1 produces sharp, vibrant, and polished images compared to v1.0 or LightX2V.

Step 6: Generate Images

Test your workflow with sample prompts. Adjust LoRA strength and CFG as needed. Qwen Lightning v1.1 delivers clean edges, vivid colors, and excellent text clarity in complex scenes.

This workflow is perfect for AI artists, developers, and content creators seeking high-quality results with minimal setup. Whether creating cyberpunk environments, fantasy worlds, or realistic portraits, Qwen Lightning v1.1 in ComfyUI is a reliable, professional-grade solution.

Comments 0 total

    Add comment