Moderate Simple Prompt travel and motion loras v1
Updated Moderate Prompts to fix the comma problem and 2 other small fixes
what it has
- full new animate diff prompt travel with explanation
- full motion lora use
- as low as 3.9gigs vram at 320x320 - 640x640
- 70 second generation at 32 frames 512x256 - 1024x512
- ip adapter
- simple manual masking
- tagger prompt help
what it does not have
- controlnets
- background controls
Completely Simple Prompt travel and motion loras v1
As far as i can tell, this version can do 500+ frame videos. I use 3.9 gb vram for any video base of 240x320 any frame number. The only thing that makes you use more ram is making the video bigger. I think as long as you can do a batch of 16 images at the size you are trying to use it should fit in your vram.
what it has
- full new animate diff prompt travel with explanation
- full motion lora use
- as low as 3.9gigs vram at 320x320 - 640x640
- 70 second generation at 32 frames 512x256 - 1024x512
what it does not have
- control nets
- ip adapter
- masking
- background controls
This works, but don't expect detailed, amazing videos. It has no background control so it goes wild sometimes and no IP ref to get consistent characters. Lora use helps.
Because of the lack of controls, if you have the memory, you can do larger base videos easier.
Advanced version coming soon, just found out you can mask motion loras.. who knew.
V2.0 - hotfix - fixed custom mask order,
- added negative schedule
- added mask use for IP adaptor
- added mask subject group for upscale.
- added wd tagger to auto prompt or help prompt.
- schedules now turn off entirely when disabled.
- setting a few constants now to avoid node line spam.
- many small QOL fixes
- Made the interface proper
v1.1
- added optional prompt scheduling
- many small qol fixes
added motion lora setup
Simple prompt travel
features
- 2 models
- LCM
- full text to video prompt travel
- lora/embedding use
- ipadaptor use
- very low memory needed
- fast generation
Description
V2.0
- added negative schedule
- added mask use for IP adaptor
- added mask subject group for upscale.
- added wd tagger to auto prompt or help prompt.
- schedules now turn off entirely when disabled.
- setting a few constants now to avoid node line spam.
- many small QOL fixes
- Made the interface proper
FAQ
Comments (3)
Is there a way to use this workflow without signing into a Chinese owned site to receive an API key?
Edit: Theres a specific node called BizyAir Nodes that caused that prompt to come up. Not installing that node module allows you to still use this workflow properly. Author put a fuck ton of work into this. Mad respect.
Possible for an img2vid version ?
Where can I download the materials to run this workflow?
such as Checkpoint, upscale model, 250_bounce_r64_temporal_unet.safetensors, 250_sideblow(20fps-5122)_r64_temporal_unet.safetensors, and 350_missionarywide10_r64_spatial_unet.safetensors
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.