Scheduled Maintenance

We're performing scheduled maintenance. Some features may be temporarily unavailable.

Choose Theme

JollyAI Default

Clean and professional dark theme

It Comes at Night New

Ultra dark JollyAI theme

Stranger Things

80s horror-sci-fi aesthetic

Batman

Dark knight, dark theme

Barbie

Bright and colorful Barbie theme

Ocean Blue

Twitter-inspired blue theme

Midnight Purple

Catppuccin-inspired purple theme

Forest Green

Matrix-style green theme

Crimson Red

Warm red-based dark theme

Amber Glow

Warm amber/orange theme

Dracula

Popular gothic-inspired theme

Monokai

Classic Sublime Text theme

Nord

Arctic-inspired cool theme

Gruvbox

Community favorite warm theme

Solarized Dark

Carefully calibrated for readability

One Dark Pro

Popular VS Code theme
2 Free Daily Trials · No Sign Up to Try · Unlimited on Pro

Free WAN First & Last Frame Video Generator Upload Two Images — Wan 2.2 Morphs Between Them

WAN First & Last Frame video generation on JollyAI turns two still images into one smooth 8-second HD video. Upload a start image and an end image — Wan 2.2 generates every frame of motion in between, with natural transitions, consistent identity and cinematic quality. No local GPU, no ComfyUI setup required.

What is first frame / last frame (FFLF) video generation?

Most AI video tools only take one image and guess where the motion goes. First frame / last frame (FFLF) generation gives the AI both ends of the motion — a start frame and an end frame — so it only has to fill in what happens in between. The result is far more controllable: you decide exactly how the clip begins and exactly how it ends, and Wan 2.2 handles the realistic motion, camera movement and transition in between.

Built on Wan 2.2's dual-pass (high-noise + low-noise) video architecture with CLIP Vision conditioning on both images, then finished with RIFE frame interpolation for smooth 32fps playback.

Why use WAN First & Last Frame on JollyAI?

Two-Image Control

Upload exactly a start image and an end image — you control both ends of the video.

Wan 2.2 Motion

Dual-pass high/low-noise sampling for smooth, realistic in-between motion.

RIFE Smoothing

Frame interpolation upsamples motion from 16fps to 32fps for fluid playback.

Last Frame Included

The generated last frame is auto-extracted as a downloadable image, just like WAN image-to-video.

No GPU Required

Runs on our servers. No Nvidia card, no local setup, no ComfyUI needed.

Unlimited on Pro

2 free trials daily on any account. Upgrade to Pro to remove the daily cap.

How to generate a first/last frame video

  1. Upload your start image — the first frame of the video.
  2. Upload your end image — the last frame of the video. Same subject, different pose/scene works best.
  3. Describe the transition — e.g. "smooth turn, natural pose change, cinematic motion".
  4. Generate and download — an 8-second HD video, plus the extracted last frame as a downloadable image.

Popular use cases

  • Outfit or look transitions for fashion and lookbook content
  • Before/after and transformation reveals
  • Smooth scene-to-scene transitions for Shorts and Reels
  • Product or pose changes without reshooting video
  • Storyboard keyframes brought to life as motion
  • Character pose-to-pose animation

WAN First & Last Frame — FAQ

What is a first frame / last frame AI video generator?

You upload two images — a start (first) frame and an end (last) frame. Wan 2.2 generates every frame in between, producing a smooth video that morphs from the first image into the second.

Is the WAN first/last frame generator free?

Yes — every account gets 2 free trials per day, no sign up needed to try it. Upgrade to Pro for unlimited daily generations.

How long is the generated video?

A fixed 8 seconds, rendered at 16fps and smoothed to 32fps with RIFE frame interpolation for fluid motion.

Can I download the last frame as an image too?

Yes — the generated last frame is automatically extracted and offered as a downloadable image alongside the video, the same way WAN image-to-video works.

What images work best?

Use two clear images of the same subject — a start pose/scene and an end pose/scene. Similar framing and lighting between the two gives the smoothest results.

Do I need a GPU to use this?

No. JollyAI runs Wan 2.2 on our own GPU infrastructure. Just a browser — no ComfyUI, no Python, no VRAM required.

Is my data private?

Yes — uploads and outputs are auto-deleted after 24 hours. Cloudflare-protected.

Ready to morph two images into a video?

2 free trials daily. No sign up to try. Unlimited on Pro.

Privacy Auto-deleted in 24h · Protected by Cloudflare · No data shared