A little late to the LTX-2 party, but finally tried it out.
Animated shape layers with keyframed warping in AE, then ran it through ComfyUI.
Love how fast it generates, and it’s especially solid if you want more precise, controlled movement like animating on beat.
Topaz Labs' flagship upscale models are now available in ComfyUI!
Video upscale - 4K
- Astra: Diffusion creative upscale (recommended)
- Starlight Fast: Diffusion precision upscale
- Apollo: Frame interpolation
Image enhancement - 8K
- Bloom: Image upscale and face enhancement
Widely used for film production, professional photography, and design.
Try it on Comfy Cloud: https://t.co/tci2fZ7KbF
Use Veo 3.1 in invideo to create continuous long videos with consistent characters seamlessly.
Veo’s first frame / last frame magic, paired with invideo’s autonomous agent that creates beautiful, narrative driven ads.
PS: this ad is inspired from Old Spice’s iconic commercial. Linked in the next tweet.
Sora 2 blocks input images of real people, making character consistency impossible.
I built a workaround so you can have consistent AI UGC actors across videos.
Product photo + Prompt → Sora 2 vid → Watermark removal → Character + Wan Animate
Step by step guide below ⬇️
これディープフェイクって信じられないな。
ㅤ
自撮り動画を入れるだけで表情・口パクも馴染む。
PR動画やVtuber風の撮影もいけそう:
ㅤ
🚩やり方
① Wan aiにアクセス
② 自分の動画と参照したい顔の画像をアップ
③ Avatar + Character Swapを選択
④ あとは生成するだけ
Calling all marketers —
Lovart × Nano Banana is your new SUPERPOWER!
Drag & drop assets. Doodle poses -> Product demos. Merch designs. Campaign visuals.
Ship social + ad-ready content in seconds.
Upgrade TODAY for 365 days of FREE UNLIMITED Nano Banana! #Lovart#NanoBanana