sabowsla/wan21-flf2v-full-quality

Wan2.1 FLF2V - First & Last Frame to Video Generator

Generate high-quality videos from first and last frame images combined with text prompts using the Wan2.1-FLF2V-14B-720P model.

Features

• —🎬 Generate videos from first and last frame images
• —✍️ Text prompt support with prompt enhancement
• —⚙️ Advanced controls: diffusion steps, guidance scale, shift scale, seed, solver selection
• —🌍 Multi-language prompt enhancement (Chinese/English)
• —📐 Multiple resolution options (720P, 1280x720, 480P, 832x480)
• —🎯 Customizable frame count (17-81 frames, must be 4n+1)
• —🚀 ZeroGPU support for efficient GPU usage

Usage

1. 1.Wait for the model to load automatically (first time may take a few minutes)
2. 2.Upload first and last frame images
3. 3.Enter a text prompt describing the video
4. 4.(Optional) Click "Enhance Prompt" to improve your prompt
5. 5.Adjust advanced options if needed
6. 6.Click "Generate Video" to create your video

Model

This space uses the Wan-AI/Wan2.1-FLF2V-14B-720P model from Hugging Face Hub.

License

Apache 2.0