Loras for vid 2 vid<a target="_blank" rel="ugc" href="https://huggingface.co/Alissonerdx/Minimax-H3-ComfyUI/tree/main/loras">https://huggingface.co/Alissonerdx/Minimax-H3-ComfyUI/tree/main/loras</a>Char swap 1.1 for replace, motion transfer tara allows replace background5.2 - Same download (small fix) Reuploaded to reconnect the audio, gahhh i need testers lol.<ul><li>Fixed ref video length for extends</li></ul>5.11 fixedFrom my tests Video replacement works best at 124 frame batch counts.V5.1- added video ref options- added instructions to do video to video replace.- added passthrough for upscale<ul><li>added kitchen attention back<hr /></li></ul>V5 - The fixing of the comfy ui updates... and probably my mistakes too...I think i fixed it all... butComfy changed the way many things worked without telling me.. so i tried to fix it...I dislike the new comfy... stop breaking every god damn workflow every update.This is a big flow... its really hard for me to test everything properly with my one little mid range PC. If you have problems, comment. I need to know if there are problems.- Added Model, clip, vae loading subgraphs- Added proper upscale model switch to allow you to use a different model/loras for upscale- Added final vae to change vae for final output.- Added full lora sets for both base and upscale- added Image ref subgraph- Changed bypasses for refs to mutes(comfy changed how bypassed gets work)- Changed upscalers to give options to set size or set upscale multiple.- added passthrough for extenders so you can pass the first batch for upscaling- added proper memories with latent cuts, fixed passthroughs with default cut- Added sound preview for batch for extending to allow proper prompting- Added Sigma change options for extending- added merge check for upscale.- Changed staged loras to subgraph- Added upscale append/appendmasks options- Defaults are 0New distilled and upscale lora options<a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3-experimental/tree/main">https://huggingface.co/Kijai/MiniMax-H3-experimental/tree/main</a><a target="_blank" rel="ugc" href="https://huggingface.co/Comfy-Org/MiniMax-H3/tree/main/loras">https://huggingface.co/Comfy-Org/MiniMax-H3/tree/main/loras</a><a target="_blank" rel="ugc" href="https://huggingface.co/Pika-Labs/Pika-H3-Flex-Turbo-LoRA/tree/main">https://huggingface.co/Pika-Labs/Pika-H3-Flex-Turbo-LoRA/tree/main</a><hr />4.11 (same download) - Didnt want to leave people hanging.<ul><li>Memory nodes need to have separate av latent unbypassed and placed outside the group.</li></ul>- Added append masks to options for the upscale- V5 will complete the append by allowing the extend to see the append frames.- Changed rtx and latent to 2x upscale. They stretch the video when its not exact. 2x seems to work much better. You can change this and set the size if you know what will work with a resolution selecter, but 1mp skews the video when upscaling from 0.3 and 0.98 skews the other way. Warning, 2.0 works. I dont know about other scales.- Resize is there with crop/pad if needed to get around this but you need to pixel upscale- Changed memories to latent cuts- added memories to upscale<hr />V4.1 fixes (Its the same download, you know you got the right one because its named 4.1)- Fixed minimax audio (audio mask fixed to match seam) Custom audio unchanged.- Changed to empty audio latent, added audio frames option- set audio frames higher than you need (even max), you dont need to use all the frames.- fixed memories to default to last frame when you bypass one.- Lightened upscale append mask- Upscales suck... I have lightend the append mask to 1 of 17 atempting to join properly allowing it to render further and not abruptly snap back to the original latent.This works i think, but that may be wrong again. My head hurts from this debugging.I still think whoever came up with this latent system should be shot.... in the knee... with an arrow...<hr />V4 - Optimizations and ease of use- Brought back partial daisy chaining so the flow moves all the way through the flow- Saved latents only needed when extending further than 2 batches.- Proper explanation notes on the 4 groups use and chain.- added appending to upscale to hold seams properly.- Added Beta57 8,6,4 step sigma values as options- Added main switches to easily turn things on/off- fixed small bugs.<hr />v3.1 - fixes --- really getting sick of breaking my own stuff fixing stuff...- Fixed preview audio for base video extending. Pretty sure the rest of the base generation worked flawlessly.- Fixed upscale latent overlap-- this was my fault for pushing out a fix for another issue before testing (sorry)-- Full issue was overlap latents were not inserted so seams were not carried over causing clear changes on batch change to elements in the video when upscaling-- This should be seamless now with testing done on multiple batch counts and all 8 batches tested.- Few small interface changes- The first upscale extender has (MTB nodes) audio duration node they are not needed for the flow to work, i forgot to remove testing stuff from only the first extend group for upscale. The audio cut can be done differently and i was experimenting. You can just remove the MTB nodes in the subgraph and you shouldnt need that set for the rest.<hr />Full minimax V3 - changes- Changed to full latent batching. Removed all audio timestamps.- Frame perfect audio both generated and loaded- Latent only passthroughs for easier chaining- Added an actual interface with switches for everything- Cleaned up inputs and outputs for easier chaining.This should now be much easier to use.Bugs to be fixed eventually<ul><li>Preview of the main renders for the extend groups audio is messed up</li><li>To fix inside the extend groups subtgraph is a trim node near the output. The division needs to be swapped. a/b not b/a</li><li>Fix to extending upscale that i dont know if is in the live version that mixes up the overlap</li></ul><hr />Full V2 -Fixes- Audio masks for generated audio now work to extend.- Bypasses now work properly- Previews now show only one batch-- Enabling any previews forces the flow to decode a segment. It can slow the process.-- Because we are no longer using the video output of the subgraphs It never needs to go out of latent space. Enabling the previews forces the flow to decode the segment.- Latents are now decoded all as one.-- (thx obvom) Batched vae decodes cause temporal issues on the seams.-- Workflow now uses the full latent video for decoding, and encoding.I learned so much.. i think my vace flows could do much of this.<hr />FullThis is ref to video with extending and upscaling- Ref to video has-- 6 Images, 3 audio ref inputs.-- 3 batch groups that can be daisy chained further if needed-- Custom batch counts with syncronised audio either generated or loaded.-- Loras, prompts, seeds, refs can all be changed per batch.Upscaling- This could be its own flow, it does not need the base generation, you can load any video or even saved latent in the upscale section if you want.-- 3 upscale options, RTX, latent, or resize-- Loras, prompts, seeds, refs can all be changed per batch.-- Auto batching<hr />Vid 2 vidThis is an addition to my extending flowMany of the options are the same.This includes a video loader and allows you to batch copy videosIncludes memory insertionThis is a beta because, i cant test everything with it. If it breaks, let me know i will try to fix it.All three extending flows now are in this version with memories options.<hr />ExtendingI made this originally to prove to myself that latent extending still contrast shifts. But i never can just stop there.Latent extending can work well, but its not perfect.Also this works off frames... not time. I got a headache with the audio/timing problems that came with the math needed for working with time. Minimax needs certain frame counts, allowing you to input seconds was useless because i had to convert all times to frame counts anyway. 10 seconds is not 10 seconds for minimax... its always a bit more or cuts and its less. This flow skips that and just asks for the frames. It will just limit your frames and overlaps to what minimax can do properly. No asking for 5 seconds and getting 6 or 4...This comes with 2 flows1 generates the audio with minimax1 loads an audio file you already have to sync, music, talking etc.I want to give a shoutout to obvpm for his nodes and flowsThis guy gets it, its nice to see someone trying to make hard things easier :)<a target="_blank" rel="ugc" href="https://github.com/chanon/comfyui-obvpm-timeline/">https://github.com/chanon/comfyui-obvpm-timeline/</a>These thoughts are basicly mine... he knows what i know about this extending stuff. I really like this guy. Give him a sub, like, comment. He deserves it.<div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/GiJxlWOooyo?rel=1" start="0"></iframe></div><hr />Upscaling v2.CORRECTION - The frames in last batch number is wrong its telling how many are missing from the last batch. Keep this as low as you can. You can leave it like that just know what it means, or you can fix it with.<ul><li>To fix create math node a = batchcount b = the old calculation int. this will give the last batch count.</li></ul>This was to help people with lesser systems.Load a video you already have, add refs, prompts and upscale in batches.Two ways to use this<ol><li>Batched like default</li><li>daisy chained to do it all in one go.</li></ol>Batched should allow anyone to upscale a video.Daisy chain is for people who have the system to do it all.Minimax can make 30+ second 0.2-0.4 MP videos, This will upscale it to whatever size you can run for the batch size you select.<ul><li>Added a fix to the final size. It should now output the same length video you put in.</li></ul>Removed all base video generations this is a bring your own video flow.<hr />UpscalingUploaded a fix and reworked a few things to explain better how to start.This can help remove contrast drift from extending videos somewhat.This was made to upscale minimax videos in batches to allow smaller longer videos be upscaled to higher res.Can make a video or load a video you already have.<hr />V3Advanced minimax- generate videos from ref images.- advanced extend videos- use custom audioincludes a beta flow for face fixing that allows batching.<hr />Minimax face fixing (inpaint)This is my old vace workflow edited to use minimax instead.Old vace flow - <a target="_blank" rel="ugc" href="https://civarchive.com/models/2570937/wan-vace-21-flow-archive?modelVersionId=2894367">https://civarchive.com/models/2570937/wan-vace-21-flow-archive?modelVersionId=2894367</a>- Cuts out faces to inpaint.- Can keep/change/create audioEnjoy :)<hr />V2Simple minimax- generate videos from ref images.- Simple Extend videos- Beta memories for extending- use custom audioI dont think i explain everything well enough, if any questions ask in comments i will add any answers to the next flow.<hr />Best ref2va distilled lora<a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3-experimental/tree/main">https://huggingface.co/Kijai/MiniMax-H3-experimental/tree/main</a><a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/MiniMax-H3-experimental/blob/main/loras/MiniMax-H3-Ref2VA-Acc-8Step_pruned_comfy.safetensors">MiniMax-H3-Ref2VA-Acc-8Step_pruned_comfy.safetensors</a>Strength 0.4-1.0I debated releasing this as its not much more than the basic comfy workflowWhy use minimax- Great prompt listening- Combine multiple subjects easily.- Quality lip sync and voice clone- 5-15 second video clips 300 frames at 1MP.- You want to create a story.This uses A LOT of resources, crashed a few times, soft locked too when i went too big, but 1MP is not a tiny video when i ran about the same for LTX and much smaller for vace. Its slow, but its doing more than most.I think it can extend like vace could, and without a contrast shift. Its down to learning the prompting to extend from the last 8 frames of a video.Minimax is a beast for storytellers.(fyi my gpu is stuck on PCI3 atm due to my cpu)From clean load of comfy loading models and render of 1MP 311 frames at 8 steps took with my setup 23 min 21 seconds 158s/iOutput is 1376x768 at 12.75 seconds

Minimax.. simple

V5.1- added video ref options- added instructions to do video to video replace.- added passthrough for upscale

Files with SHA256: ffc0697857ce0e3bd4528ac39c9d00fd0bc5b3b8d18f8e28a223b4bdb1b42217