Seedance has a hidden feature that feels illegal. It lets you create old-school anime at lightning speed. It’s called “in-between” - and it actually delivers. Try it yourself 👇

This video took me 4 hours to make using this method. But speed is not that important. STORY, DIRECTION, CONSISTENCY, PACING and your overall taste is what really matters. Focus on those - otherwise you'll end up with just another random AI clip.

Step 1: Find an image that sparks your imagination. The more inspired you feel, the higher the chances you’ll create something cool I picked this 👇 because of its dark, oldschool “Akira” anime vibe (you can steal this style by prompting NB: “Create X in this style")

Step 2: This is where "in-between" technique starts. Upload this pic to any platform (I use @Morphic) and ask Nano Banana (Pro or NB2) to: "Show what happens in 5 minutes. Soldiers stand by the enemy military base gates"

Step 3: Use these two images as the FIRST and LAST frames in Seedance 2.0 to generate 10-15s video. Prompt (MOST IMPORTANT PART): “Show what happens in between. Soldiers run through the snow towards a military base. 5 different camera angles. No music” This prompt structure is the key to making the technique work. Keep in mind: “Show what happens in between” and “5 different camera angles” are the DEFAULT parts. The sentence in the middle is the CUSTOM part - that’s where you describe exactly what you want Seedance to generate. (in my case it was "Soldiers run through the snow towards a military base.")

Not every shot the model generates will be perfect - but it will do the job. It will show the soldiers approaching the enemy base from 5 different angles. (To improve the final result, generate a few more versions, pick the best shots, and stitch them together in your editor.)

Step 4: Here is where you simply repeat the process: 1. Re-upload the previous LAST frame to Seedance as FIRST frame 2. Use Nano Banana to generate the next LAST frame (in my case soldiers hiding from security guards)

Step 6: Upload both images to SD again and use the same prompt structure - just change the middle sentence based on your story. “Show what happens in between. Soldiers enter the base, run through narrow corridors, and hide from guards. 5 different camera angles. No music.”

Step 7–12: Repeat the same process: 1. Move the LAST frame to the FIRST frame slot for each new generation. 2. Generate a new LAST frame with NB. 3. Reuse the same prompt structure, only changing the middle sentence based on your needs.

Also remember - for the best results, generate at least 3–4 versions of each sequence, pick the best shots, and combine them in your editor. (That’s why this took me 4 hours, not 30 minutes.)

The old anime animation will be part of Cartoon Hero 3.0. Inside, I’ll show you step-by-step how to create it and master the “in-between” technique. The course drops in 9 days. Join the waitlist 👇

I’ve been saying AI digital “in betweeners” will bring back old-school animation. Artist come up with the key frames and most of the animation, AI fills in the gaps.

yes this is completely true!

In-betweening used to eat weeks of an animator's life, so watching a tool handle it instantly is wild.

indeed:)

This will save alot of time. Thanks for sharing, Framer!💫

Try it, want to see what you come up with!

Amazing, lots of animators literally would want this to be illegal. They didn’t complain when digital animation replaced hand drawn, but now that its their job on the line they want to pull up the ladder for anyone else

That's excellent mate, nicely done.

thanks buddy:)

Thank you Sensei, will give it a try, I’m more or a donor shot and best frame guy but I definitely see how this keeps consistency

but does it work on seedance? i used donor shot in sora 2 only, now swtiched all the prompts and model to sd in the course

It handles it well, the whole Animan's world series is built using the format. For continuity I use the donor shot and last frame together.

@Framer_X what is donor shot ?

@Framer_X It’s a shot that contains critical elements of what story you want to tell e.g environment and the characters made some props as well.

This came out so good, Framer. I love this art style and overall scene 🔥 Were the SFX and Music always generated in Seedance or did you add those after?

thanks! always added later in video editor. If you need something longer than 15s you need to prompt SD "no music" so you could later use your own

Absolutely thank you, Framer! 🙏 I love your work and I really appreciate you consistently providing resources for the creator community! You're a legend 🙌

music for my ears:))

Perhaps we will finally have proper Claymore Season 2 soon

Pretty nice workflow! Would love to implement it to create more content in short time!

try it, lets see what you create:)

Awesome stuff as always sensei

haha thanks man

Your tutorials are always a goldmine ! Thank you Framer as always ❤️

appreciate your comment!

저도 아키라 스타일 좋아하는데 대체 어떻게 만드셨나요? nice !

you can generate this style with ai, or just take my image and ask ai to generate any other scene with your characters using my style )

Thanks for sharing this.

always for your service ser

Thank you I think this is where I can really make something cool now

amazing! hope you will use it in your youtube journey!

This is amazing

Wow

Great workflow Framer 👏

so at lightning speed = 4 hours

yes this is very fast. compare to traditional animation, would have taken 1 month for 5 ppl and 20k

Seedance too pricey for me to try :(

go with 480p (cheapest), also can try runway unlimited plan 99usd month, allows to generate 2 videos at once, though its generates quite slowly (20mins for 1 vid)

Cheapest I saw was 30 a month. I was hoping to have a trial of sorts before throwing that around. Am a very broke Koji

That's amazing. Thanks for posting...

your welcome buddy

The no music prompt is indeed a pro tip. Seedance tends to do that too much.

"...lets you create..." Dude. You created shit. AI created.

Very useful, thanks Framer

Pretty interesting, thanks

@Ludedwolf

Do this to create breast expansion clips 🧐

So— in this case, the person directed nothing. This is an impressive technology demo, but not an actual example of someone creating purposeful art with these tools. It’s like me rendering some basic Blender sample scenes and posting them. Do *real* things with the tools.

nice stuff

thanks

Thank you! really help. the output is fire tho!

@huykao_ai 🫡

Living in a time where this is possible and not having any time for creative work makes me want to cry

oh sorry to hear that. hopefully it will change!

nailed it

Genial workflow as always!

@stonekaiju Awesome stealth video workflow.. Thank you for sharing the process 🙏✨

its insane