WAN 2.1 FLF2V: Simple ComfyUI Workflow for First-Last Frame Video Generation (Low VRAM Friendly, Laptop)

🌀 WAN 2.1 FLF2V: Simple ComfyUI Workflow for First-Last Frame Video Generation (Low VRAM Friendly)

This guide introduces a straightforward ComfyUI workflow for generating videos using the WAN 2.1 First-Last Frame-to-Video (FLF2V) model. Designed for systems with limited VRAM, this setup enables the creation of smooth, cinematic transitions between two images ideal for storytelling, before-and-after showcases, or creative animations.

📥 Required Models and Files

Place the following files in their respective directories within your ComfyUI installation:

• WAN FLF2V GGUF Model: wan2.1-flf2v-14b-720p-Q5_0.gguf → ComfyUI/models/unet/

• Text Encoder: umt5_xxl_fp8_e4m3fn_scaled.safetensors → ComfyUI/models/text_encoders/

• Clip Vision Model: clip_vision_h.safetensors → ComfyUI/models/clip_vision/

• VAE Model: wan_2.1_vae.safetensors → ComfyUI/models/vae/

Note: If you have less VRAM, consider using lower quantization versions like Q6 or Q7 to reduce memory usage. check out the Links in my other workflow.\

https://civitai.com/models/1309674/simple-wan-21-low-vram-comfy-ui-workflow-gguf-4gb-vram-16gb-ram

🧰 Workflow Setup

1. Update ComfyUI: Ensure you have the latest version of ComfyUI installed.

2. Install Necessary Nodes: Use the ComfyUI Manager to install any missing nodes required for the workflow.

3. Load the Workflow: Import the provided JSON workflow file into ComfyUI.

4. Input Frames: Drag and drop your starting and ending images onto the designated nodes in the workflow.

5. Configure Settings:

   • Resolution: Set according to your desired output (e.g., 720p).

   • Frame Count: Determine the number of frames for the transition (e.g., 81 frames for ~5 seconds at 16 fps).

   • Prompts: Input positive and negative prompts to guide the video generation.

6. Run the Workflow: Execute the workflow and wait for the video to be generated.

🎥 Demonstration Video

For a visual demonstration of the workflow, refer to the following video:

Note: While the video showcases a different workflow, the principles and setup are similar and can assist you in understanding the process.

🧠 Reasoning and Tips

• Model Selection: Choosing the appropriate quantization level (e.g., Q5, Q6) balances quality and VRAM usage.

• Prompt Crafting: Well-defined prompts enhance the coherence and quality of the generated video.

• Resource Management: Monitor your system's VRAM and RAM usage to prevent crashes. Adjust settings as needed.

• Experimentation: Feel free to tweak settings like frame count and resolution to achieve desired results.

By following this guide, you can efficiently generate videos that transition smoothly from a starting image to an ending image using the WAN 2.1 FLF2V model within ComfyUI. This workflow is optimized for systems with limited VRAM, making it accessible for a wide range of users.