Wan 2.1 FLF2V 720p F16
Generate videos by controlling first and last frames using Wan 2.1 FLF2V.
About
This ComfyUI tutorial shows you how to generate smooth, guided video clips by explicitly controlling the first and last frames using Wan 2.1 FLF2V. The core of the workflow is the WanFirstLastFrameToVideo node, which fuses three condition sources: your prompt embeddings from CLIPTextEncode, visual guidance from CLIPVisionEncode of the start/end images, and the Wan 2.1 UNet loaded by UNETLoader. The workflow is configured for 720p output and 16 frames (F16), giving you a compact, visually coherent clip that cleanly transitions from the first frame to the last.
What you provide
- Reading the workflow
Runs in ComfyUI. This workflow can't run on this page. Open it in Comfy Cloud with everything loaded, or download it for your own ComfyUI.
Try in CloudQuestions & answers
Yes—this FLF2V workflow is designed to use both. If you only have one image, duplicate it into both inputs to produce subtle motion without a large scene change.
Increase the frame count parameter (e.g., F24 or F32) to lengthen the clip, and adjust the resolution fields to your target size. Keep aspect ratios consistent with your start/end images to avoid padding or distortions. Higher frame counts and resolutions require more VRAM.
Your start/end images and prompt coherence matter most. Technically, KSampler steps and CFG scale influence stability: slightly higher steps and moderate CFG (3–6) often improve consistency. If exposed, motion or guidance weights in WanFirstLastFrameToVideo can also tighten the match to the keyframes.
Use a strong negative prompt (the provided Chinese example helps), lower CFG slightly, ensure both images share exposure/composition, and raise steps moderately. Also verify models (UNet/CLIP/CLIPVision/VAE) match the Wan 2.1 repackaged set and that resolution aligns with your inputs.