⛩️💀 DaSiWa MiniMax H3 💀⛩️
My new MiniMax H3 model.
Version overview: https://civarchive.com/articles/23495/dasiwa-model-versions-and-timeline
⚠️ Make sure to open the DOWNLOAD dropdown to see all quants possible.
🔮 Key Features
See Announcements!
⚠️ What NOT to expect
No tuning for ToS-violating concepts
Not more unaligned/altering boundaries/breaking guardrails
🍒Workflow
Make sure to checkout my easy to use Workflows!
🍄LoRA's
But: This checkpoint is not meant to replace LoRAs, it is meant to:
Perform better overall at his own
As easy as possible to use
With LoRAs to be more awesome
🛠️ Recommended Settings
Non-Distilled
res_mulistep/simple or euler/simple
Shift Video 10-12
Shift Audio 3-5
20-25 Steps
Turbo (Distilled)
euler/simple or LCM/simple
Shift Video 6-12
Shift Audio 3-5
4 or 8 Steps
Dependencies
See Workfow notes.
🩻 Known issues
Tell me 🫵🫢
🩺 Fixes & Feedback
Update your ComfyUI ❗
🖤 Why I Made This
Pushing MiniMax H3 to its limits!
This checkpoint is also my personal playground.
Closing words
🤩 I want to thank all the fantastic other creators who made super nice LoRAs and concepts to play with! Support that awesome creators by using their LoRAs and post to their gallery and share the meta-data!
⚠️ I made all this with permissions or open-source resources (the time it is incorporated).
I share as much insights as I can without compromising my work. I'm doing this for fun as my hobby and just do not want my hobby to be destroyed.
More details can be obtained in the corresponding announcements!
If you would like to contribute in my awesome (😉) checkpoint or willing to share resources I'll gladly give credit! Just contact me!
✅ All credits / resources are mentioned inside the announcements! - Since different versions may have different resources.
YOU are responsible for outputs as always! If you make ToS violating content and I get aware I WILL report this.
🔐 Official Permission from MiniMax
In August 2026, MiniMax granted me written authorization to use MiniMax H3 and MiniMax H3 Works via my license request. This authorization is personal to me as the author of this model - it does not relieve you of your own responsibility under the MiniMax H3 community License Agreement and its Acceptable Use Policy.
⚖ Disclaimer & User Responsibility
This model is a community-quantized derivative of MiniMax H3, distributed under the MiniMax H3 Community License Agreement (see the license on this page). By downloading or using this file, you accept and are responsible for compliance with those terms, including the Acceptable Use Policy and the laws of your own jurisdiction.
All outputs, and all use or sharing of this file, are your sole responsibility. Outputs are machine-generated and must be disclosed as such when published publicly (AUP §12). I do not support or take responsibility for illegal, harmful, or harassing uses - and I will report content I become aware of that violates the license, the AUP, or platform terms of service.
THIS FILE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND.
I am not liable for any damages arising from its use. I am not affiliated with MiniMax.
If this authorization is revoked, or MiniMax objects to this community redistribution, the files and this post will be removed promptly upon notice.
📜 DaSiWa Custom Addendum: Fine-Tune Integrity & Attribution
Base license: MiniMax H3 Community License Agreement -
https://huggingface.co/MiniMaxAI/MiniMax-H3/blob/main/LICENSE
This is a general-purpose community derivative. It is not a specialized or task-specific variant; its capabilities and restrictions are those of the underlying model and its license.
1. Verification & Integrity
This checkpoint is a quantized / merged / finetuned derivative of MiniMax H3. Notice of Non-Support: copies hosted on third-party platforms (mirrors) are considered Unverified. I provide no warranty, support, or safety guarantees for unverified files. Per license §III, redistributors must keep the embedded metadata, recipe files, and license notice intact.
2. Trademark & Branding
The name "DaSiWa" and associated logos or promotional imagery are the creator's i intellectual property.
3. Commercial Use
Per license §IV.1, commercial products or services generating more than USD 20 million in yearly revenue require a separate prior written authorization from MiniMax; commercial products must prominently display "MiniMax H3" (§IV.2). Using this model's official branding to market a paid or ad-supported generation service requires a separate agreement with the creator.
Description
Turbo - 4 step
FAQ
Comments (164)
Have there been any updates that would require me to re-download the Non-Distilled version?
The REF2VA quality for 4 steps is insane. How did you do this?
the turbo model？ still downloading the model, looking forward to the results!
@azsaber Yeah, this was my first test with just 4 steps euler/basic:
https://civitai.red/posts/30675099
There is some smudging, but I did it at .52 MP and 4 steps which is crazy. Bumping up the MP or steps to 6-8 will be my next test to see the generation time vs value add. I am poor and on a 3060 12g and the linked 10 second clip took around 8 minutes to generate.
Black magic!🧙♂️
@Darksidewalker CLEARLY
@FirstPrinciples Actually it is a blend of distillation, I tuned to fit from the ltx turbo loras.
Most of the time it works flawless, but depending on extra loras or latent-upscale it can overbake. But yeah, nothing is perfect. But hell yeah, 4-step high fidelity outputs!
is there a chance for a gguf models ?
INT8 CR is already AMD and NV 20/30/40/50xx compatible and much better, is there a reason for gguf?
i have only 16 gb vram
@bartoszpaprotny1997806 The model works with 12+ GB VRAM, like the normal int8 base model
"I shot this with a 0.7-megapixel setup — 5-second clip, 20 steps total, which took about 250 seconds to render. Then I tried the 4-step acceleration LoRA, and it finished in just 87 seconds. Honestly, you can barely tell the difference in visual quality between the two — the 4-step version looks just as good as the full 20-step render. The only catch? The audio from the 4-step LoRA is basically nothing but harsh static. But visually? Absolutely killer."
The released of the v1.1 turbo 4 step 768 was a gae changer for turbo lora FL2VA, 8 days ago.
If you are still not using it yet for all your T2V, I2V, FLF2V, you are missing an amazing change in term of quality and speed. and just like to waste your time :p
An personnaly i found the way to get the exact same video /audio quality for R2V at 4 steps, even at the lowest resolution 0.20 mp it's amazing quality.
But as it was a long testing phase for few days i will keep it secret for now.
@Pat3dx Thanks, I'm gonna go give it a shot right now.
@Zh_an https://huggingface.co/lightx2v/Minimax-h3-Turbo/tree/main
Don't forget to use 1.3 strengh for the v1.1 turbo 4 step lora, this is very important.
Shift video/audio 6/3, and Euler/Simple.
don't work for R2V.
@Pat3dx this lora is still giving me smeary visuals at higher res and garbage metallic audio.
for the checkpoint (turbo) you need other settings than for normal, or you will have issues.
@SweetAyanna Strange, i have no issue and perfect quality, even at 0.20 mp. perfet video and audio. something must be wrong in your workflow.
@Pat3dx We have extensively tested most turbo loras in the official Comfy Org discord server. And not a single individual that I've seen has gotten any good results with that v1.1 lora. And no there is nothing wrong with my workflow lol. You're the first case that claims to get good results with this. If you want to prove me wrong however you can link a generation with that exact lora, with SFX + dialogue in it. meta data attached so I can verify.
But to actually add to this discussion. When talking 4 steps loras, they are always inferior when it comes to visual and audio (atleast in H3 so far). But the absolute best 4 step lora right now in my opinion is still KJ's: https://huggingface.co/Kijai/MiniMax-H3_comfy/blob/main/loras/minimax_h3_fl2v_lightx2v_turbo_4step_v0.1_comfy.safetensors at 0.75 strength with er_sde/beta. And if you want the absolute best results, almost indistinguishable from base, you use this 8 step lora at 1.0 strength and 8 steps: https://huggingface.co/lightx2v/Minimax-h3-Turbo/blob/main/minimax_h3_fl2v_turbo_8step_v1.0_comfyui_bf16.safetensors
no video/audio shift needed since those work with the base shift values in comfyui.
@SweetAyanna This is the Lora i was using before i saw the results of the enw v1.1, of course. i tested all turbo lora with all checkpoint official int8, nvfp4, etc...or fine tuned. Eros, remix, and pinkcherry.
but the v1.1 just destroy it 10x better.
check my last video posted for the Penislora this afternoon.
https://civitai.red/posts/30670172
Resolution 384x544 - 4 steps, using the new turbo 4 steps v1.1 768 lora at strengh 1.3 - 40s video generated in 387s with my 4090.
And this is just an example. As i'm making like 100+ tests everyday while i'm developping and enhancing my custom workflow/custom nodes.
giving you one of my video will not help you, draging n dropping my video will just load a full red workflow, as i never use native comfyui node, and always develop my own. So, all my workflow are fully custom, no native comfyui node in any of them.
@Pat3dx That example shows the exact issues with that lora. Smeary/Ghosty motion. Bad/Metallic audio. And 387s for that res is extremely long even for 40 second gen on a 4090.
@SweetAyanna i don't used my dual sampler with latent upscaler. Just my normal H3 workflow single pass.
With my new dual latent upscale, it's 3-4x faster of course.
And trust me i tested all lora with same seed, prompt, with all lora. The difference between the new v1.1 FL2VA turbo lora and older turbo lora, is huge.
You will never get this quality at this low res, NEVER.
My goal posting this video this afternoon, was not to proove and show you can get production ready/excellent quality with the new v1.1 lora, of course not, but to show how good is the quality compared to previous lora at this extreme low res.
and all the comunity on the Lightx2v ntocied exactly the same amazing quality this new lora provide. You may be one of the only one not getting this quality. So maybe you should make another tests with different workflow/parematers.
And what you surely not noticed, and that make this video even amazing, is that it's not an I2V, but R2V.
I used my technic to take advantadge of the new fl2va v1.1 turbo lora 4 steps, for R2V. what is sown on this video is fl2va turbo lora used for R2V video.
And trust me, with the R2V turbo lora, good luck to get this amazing quality at this extreme low res and only 4 steps.
Everyone know the FL2VA is more powerful in term of quality, motion, consitency than the R2V model. So getting this quality with R2V and previous lora, is just IMPOSSIBLE.
for a correct video quality, not production ready, consistency, you need at least 10 steps with R2V model+previous lora.
you can male all tests you want, at only 4 steps with all sampler, you will get horrible video.
大丝袜的模型，除了油腻，堪称完美。希望能去去油。
我也发现会些微改变女性的皮肤颜色变得更偏黄色和油腻。期待后续版本
is there a chance for a int4 version?
1) Acc PDD gives arror. It says that model is FL2VA
2) As Ref2VA this model bad
1) 💀
2) 💀
хз все работает идеально
@Darksidewalker This man is using a potato and blames you. Lol.
I am shocked. Why you are not asking one million dollars to download your model?
Free stuff for great art ✌️
@Darksidewalker we all appreciate it
I was wondering what happened to this model... Why did it suddenly disappear?
Looks like a licensing issue. Description suggests this was resolved.
this model without turbo lora is better and faster ,its the fastest model minimax for my workflow with ok quality good job (this model strong in image reference not in the prompt)
record time (latent upscale) 0,2 to 0,5 , 10 second =200 sec with rtx 3070 8gb, ram 16 gb.
😳
what settings are you using? oh and which version?
@mugibot72 i using the 8step turbo hybrid with latent upscaler workflow..
Your model and workflows are working very well (even though they seem quite complex to me) and are doing a great job—thank you and congratulations. I have a request: please add an on/off function to your System Monitor node. I can’t turn off the components I don’t want, and there are a lot of indicators that are completely unnecessary for me. Every time you update the node (I’m sorry to say this, but), I have to delete your System Monitor node. :)
There is a turn off inside the menu and there is a ENV you can set to completely disable it. Description is on github, as always.
is it the quality of 8 steps is better than 4 steps but lots slower? And how about the not turbo one?
I've downloaded all three but they seems are the same quality in ref2VA...? My setting 8 steps res_multistep simple shift_video12 shift_audio3. Am I confused something?
for turbo you may use euler+simple, not res_multistep.
Naturally it should be: non-turbo > 8-step > 4-step ; In quality.
@Darksidewalker Boss, one thing had to be ensured first, should I add turbo lora in those 8 steps/4 steps model? As your precious wan2.2 is lora free. And could I add turbo lora for the non-turbo version to speed up?
@biosal Turbo Model's do not need extra turbo lora's. The Non-turbo is for adding whatever you want.
@Darksidewalker Thankssssss!!!!!!!!!!!
Turbo 4 step has atrocious audio lol
Depends. But I agree sometimes the audio is not the same quality like non-turbo and 8-step. If there is any method to make 4-step distillation better, I would include that next.
use 8 to 10 steps to fix audio, its all h3 tunes
8step model is amazing, audio is great too! Did not expect it to be so much better than the original. This is brilliant work, and very much appreciated! Will you update the original (non turbo) with whatever you learned from making the turbo models?
I agree, the quality is GREAT, better than base (int8 pruned) model with 8step lora.
They all share the same enhancement, only the baked turbo and how the turbo is baked is different.
@Darksidewalker I see! it's strange then that the 8step delivers such a noticible higher output, the audio in particular took me by surprise. Once again, thanks for everything you've done, this is fantastic!
4step turbo is actually 10/10
Somehow this model makes everything darker, at last in rev2va. the previous model didnt had this problem.
Same here — all the new models show a color shift. On 10s videos the first and last frames differ quite a lot. if you extend the video to 20s, the semi‑real animation turns into a full anime style. This wasn’t present in the previous version.
+1. Despite explicit prompting of brightly lit room, the output just makes the windows brighter but the characters are very dark and lots of shadows.
I did a full 25s realistic video test with the 8turbo and no loras. There was no style shift at all.
There was no prompt given to not influence the model, that's why I posted it into my discord on the fails channel.
@Darksidewalker I testet it with a a 15 sec anime style scene, with lots of references and while vanilla h3 and your old model could match the colors, your new models made the character way darker. I testet it with the 4 step and no turbo version.
@Monorix I noticed the same, but only with the 8Step Turbo, the Standard version + Turbo Lora stays close to the reference images. Have not tested the 4step version yet.
@Darksidewalker I've got the same issue no matter the settings, everything is darker using the standard model, haven't bothered with the turbo versions.
I was doing some tests with the hybrid and 8-step hybrid models, and I'm not getting the results I was looking for.
EDIT: Sorry, I hadn't read the "What NOT to expect" part. It does not have NSFW capabilities beyond the original model. My mistake. Since I use your other models, I just assumed incorrectly.
Great work! (I'll leave it at that!)
I tested all three models, and it seems that using an external Turbo_LoRA—rather than the built-in Turbo—is the best approach for now. Of course, that depends on the use case...
i do agree, after some test, using non turbo hybrid+ external turbo lora follow my prompt better.
what turbo lora are you using?
@jgillservices377 I used "minimax_h3_turbo_v4_step600_pruned" for the version I just uploaded. However, I am still experimenting—such as by combining it with other LoRAs—to find ways to suppress noise.
When using the four-level model, a phenomenon occurs intermittently where there are two main subjects. It is like a doppelganger.
This is a feature! (Doopelganger powerrrrrrr!)
Damn, this works so well with 4 steps.
I have to say, you've always been the best!
Don't know what's the secret sauce, but the recipe is good !
thanks
I just want to say thank you for all your contributions. Please don’t let the words of a small number of disrespectful and ill-mannered people get to you. They don’t deserve your attention.
Thanks :)
@Darksidewalker weird tangent here, but if you go to ANY review on Consumer Reports, every single large appliance has ~1 star ratings because the annoyed and angry people are ALWAYS louder and more willing to voice their opinions than the satisfied... don't forget that for every 1 critic you have, there are probably 100 people using your work to see their fantasies come true... we appreciate you
@rkyha639 I try to remember that :) Thanks
WTB slave slime girls, in rl, dm...
@zugzug69 Poor sweet slimegirls! 💦
@Darksidewalker appreciate u big dawg
Hi is there a workflow instead create 30 sec long vid but also in 1 run make 3 video but seamless in 1 prompt?it Will help with low vram People like me
look up the h3 multishot node from joeygambino
You can try this plugin: kongbai-0421/ComfyUI-H3-Auto-Director: MiniMax H3 ComfyUI multi-segment auto director with multimodal references, resumable AV context, caching, and final assembly.
If your comfy breaks because of this "Director", like many others reported, do not come here to complain pls 😸
try running models through these nodes and set the chunk forward to 8-12 if low vram
model>lora>SageAttn Patch H3>Patch Sol-Attn>MiniMax H3 Fused Modulation>MiniMax H3 Chunk Feed Forward> MiniMax H3 Cache
Pls help fam this is a good model i'm having a blast with it but a 10s video takes like 30 mins to gen while a 5s vid only takes 200~s ToT i don't know what's wrong with my setup
The video memory is full and part of it goes to the RAM, which is much slower.
Are you ever going to release again a update of your pulled version. I'm doing single frame images with it, and whatever you had going on with that thing is f'n amazing for still images.
any idea why did he pull it?
Great stuff, the Ref2vid is reallly impressive. I wanted to try the turbo variants, but I only see the standard v1 available for download.
The headline show 3 variants for me
@Darksidewalker Turns out I am just stupid, I thought the 4 in the turbo model stands for 4bit integer and not for the steps, and I was confused that it stays at 8bit, and the filename was cut exactly, so that all models had the same name.
@HorstDieter The naming comes from civitai, the files actually have different names 😑
@Darksidewalker So I tried 8Step turbo version and compared it to the standard version with a turbo lora, same settings 8 steps, same 4 images as reference, tried both euler and res_multistep. The Standard+Turbo Lora really seems to follow the style of the reference images, while your 8Step turbo version, seems to ignore the style of the references images and produce a more comic/anime style. Is that intended?
The GOAT
Incredible work!
The 8Steps Turbo version causes the original art style of the anime image to be lost
but the non-Turbo version works like charm.
(Non-turbo and PDD Acc was best for me)
Thanks!
I have a nvidia 16gb vram and 32 system ram. Would this be able to run on it?
I'm trying to use your workflow. So for the hybrid, i assume I can use it for both fl2va & ref2va models. DShould it continue using the same vae and clips?
you should be able to i run it w 8gb vram and 64gb sys ram and ram only hits roughly 50%. U prob dont need to w 16gb VRAM but for me running model through these nodes makes low vram work like a charm still.
model>lora>SageAttn Patch H3>Patch Sol-Attn>MiniMax H3 Fused Modulation>MiniMax H3 Chunk Feed Forward> MiniMax H3 Cache
@goosterben483 Huh? 50% with 8gb vram and 64gb ram? I have 70-80% ram usage with 16gb vram and 64gb ram on Linux with all settings turned on in DaSiWa MMH3 v18 Workflow
@HexogenAI --disable-dynamic-vram --disable-smart-memory --cache-none --low-vram <- start comfy with those modifiers and watch the magic happen. I'm running with 8gb vram and 32gb sys ram. Just, but it fits.
@n3303422573 thanks, but i definitely won't be using those, because my system can handle all files and cache fully loaded to ram, i don't really need this. Using those arguments would only slow down the generation process for me.
How to avoid waxy/plastic skin! I am getting slightly waxy type skin texture on real human. I am using the hybrid version, no speed lora.
Are you using MysticXXX with this? If so, drop that weight to 0.2 or leave it out altogether.
Thanks for your amazing works!
I've tried both 4stepsTurbo and 8stepsTurbo version.
Your Checkpoint really does have better prompt understanding than other models!
I love the 4StepsTurbo version more than another, while it keeps the character features from reference while having good motion and prompt understanding.
I think the 8StepsTurbo version seems to have a tendency to alter the characters' facial features.
good model, tried the non-turbo variant. It works well and main improvement I see is better face consistency but it comes at the cost of losing some "details knowledge", for instance the model now seems to ignore things such as wind effects. It can do it but much less than the original model with identical prompting.
maybe lucky seeds but for any other low vram users, im getting reallllllly good results and speed using 10 steps on the 8step turbo running it through:
PatchSageAttention KJ>SageAttn Patch H3>Patch Sol-Attn>MiniMax H3 Fused Modulation>MiniMax H3 Chunk Feed Forward>MiniMax H3 Cache>Easy Cache
A little over board but gets results. Very very very mild settings on easy cache tho.
生成比较遵循提示词，但是油腻感太多了。图片参考出来的人物太油腻。4步在低步数确实不错。不要和其他lora一起用，会得到诡异的画面。
ok so for the turbo versions do we still need the turbo loras or are the built in
You don't need turbo lora for turbo version, it's already built-in
Hey, love your work. Any chance you'd upload w4a8 versions of your H3 models, or BF16s so I can make them with my converter?
Didn't want to do it for now. I have my own tool that could do it just fine.
The major point why most don't release a w4a8 for now is, that it cannot hold up on quality and people start complaining about the checkpoint and not realizing they use a ultra compressed quant that for sure have lower quality.
Maybe if some other major techs are evolved enough to compensate that.
@Darksidewalker Understandable. I will say that w4a8 is like space-magic levels of good. Where Wan2.2 Q4 was barely useable, H3 w4a8 feels like it's between 6 and 8 bits, not 4. Beats the pants off of Q4 and NVFP4, that's for sure (plus it's fast on pre-50 series cards :P) Convrot kinda is the compensating tech.
Honestly a BF16 would be plenty, from there people could make whatever quant they prefer and the poor results they get are their fault. :P
That said, yeah, logistics are a thing. I've been responsible for projects that got WAY beyond my own ability to support so I get it!!
@Darksidewalker
It doesn't have to be one or the other.
I'd recommend leaving the attention at int8, but you can still save tons of space converting the MLP layers to w4a8 without changing output quality much at all. You just need to specify which tensors get what.
...But my issue with w4a8 is that some runtimes don't seem to like it, and it can silently drop inference speeds.
@alphaatlas100844 Thats what w4a8 is doing, saving the attention layers into 8-bit, the weights to 4-bit. ... This does not magically have no impact on quality/adherence.
I added the w4a8 variant
@Darksidewalker Hell yeah, thanks so much!
I'd tried requantizing the int8 and it was pretty visibly degraded compared to going right from BF16 to w4a8, which to be fair, I did expect. xD
Hi! I really like your models and workflows. I've used the 4-step and 8-step ones and I've noticed that the camera doesn't move much. Could it be that I need to increase the shift?
work ffurry
The animation style changes in the 8‑step version, hope this can be fixed in future updates.
Just to clarify, should the 8turbo version be run with or without a speed lora?
Without, that's what distilled means, it has it baked in
Not for this theme - do you make uncensored LTX 2.5 in future?
No plan made yet
@Darksidewalker So sad... Your LTX2.3 is good. And Minimax is good.
having an odd issue, everything works except the final Enhanced video combine node. the error claims i don't have ffmpeg installed but the costom node file from the link was installed i even removed it and installed it from the extensions manager
It's there in my file but for some reason it's not using it. any help would be appreciated.
# ComfyUI Error Report ## Error Details - Node ID: 2568 - Node Type: DaSiWa_EnhancedVideoCombine - Exception Type: RuntimeError - Exception Message: RuntimeError: No FFmpeg executable was found. Install FFmpeg or imageio-ffmpeg for the mandatory H.264/MP4 fallback.
If it says you have no ffmpeg installed, you did not install it correctly.
Install and maybe restart your pc.
@Darksidewalker got it to work leaving a link that helped me in case anyone else has install issues in the future https://www.youtube.com/watch?v=eRZRXpzZfM4&t=22s
thanks for your work! Do you have any plan to make non-pruned version of fl2va or ref2va?
Actually not planned, I couldn't see any advantage over pruned. Did you have any evidence that would justify that?
@Darksidewalker I think the new VDN connection is actually only compatible with unpruned models, so pruned only misses a big speed increase from being able to do 30-step quality at 8 steps with your model. Think people still catching up to VDN tbh
@gradersayslicing847 any source that this is non pruned only?
Using non-turbo version.
Can anyone help with @nal penetration videos? It wants to merge the anatomy together no matter how clear the prompt or the image.
Could you please upload int4 convrot version of turbo models?
w4a8 8turbo is up, the 4turbo will follow
Do you know if poss to make compatible with VDN? https://github.com/OpenVDN/vdn-minimax-h3
Don't know yet
I would second this. The VDN 8-step setup is amazingly fast with 15 second videos.
I’ve been using your WAN GGUF version for quite a while now, it’s been incredibly helpful for someone like me running an RTX 4070 TI.
Today, I spent the whole day testing the Hybrid V1 int4 version for I2V with audio. I was amazed at how obedient the model is, it generates perfectly fitting audio as long as the prompt is right.
What’s even better is the flexibility to swap the standard 8-step Turbo LoRA in and out depending on my needs.
Thanks, DaSiWa. I don't know what kind of magic you used here. Wishing you all the best!
I'm excited to try this! I've had one installation issue, however. When I try to run it in REF2VA, I get this error: - Node ID: 1512:2700 - Node Type: MiniMaxH3DirectorGuide - Exception Type: TypeError - Exception Message: TypeError: MiniMaxH3ReferenceToVideo.execute() got multiple values for argument 'vae'
I read about that the comfy latest should be used, this error is not from the checkpoint, it is the comfy installation or wrong files used.
I fixed this exact issue by updating the comfyui version from current stable (or whatever it was) to latest github verison.
Thank you as always! I noticed one minor issue—is there any way to fix the slight shift in color? Other than that, it's absolutely perfect.
Me too , i got color shift and contrast change .
Hope @Darksidewalker will fix that in the future <3 .
Btw , i love the motion so much , it is perfect for character animation <3
I fixed it manually by using the Comfy integrated adjust contrast/brightness nodes before saving the video:
Contrast to 0.88
Brightness to 1.09
Those settings should be fine from what I've seen so far, it's not perfect but it's better than the extreme contrast you get by default.
Edit: If you generate videos with a very dark first frame then I don't recommend using those settings, it'll just blow the colors out.
@LastAssignment Thank you so much for your suggestion ! <3
Do you try to fix the ghosting box that was appear in the videos ?
@TheBodyHunter I haven't had any "ghosting box" appear for me, would need an example to see what you mean.
@LastAssignment Something like a transparent box and lines in the video just like when you look through glass edge
@TheBodyHunter Definitely never had that happen before.
Your WFs and models are just awesome. Really appreciated. If you are looking for missing features you maybe wanna add -> support for longer videos would be nice :) .
The model looks amazing.
I do have some burn in issues with the first few frames though but I'm not sure if its the model or just H3 doing weird things...
I'm usually getting a sharp line below the lips of my characters, there seem to be some kind of artefacting going on, something I never experienced with WAN Video.
v2 should fix that
很好用，簡直就是最棒的，但想請問一下多段拼接的時候有沒有可能多一個角色參考幀呢？
Minimax h3 is limited to the given references possible with the director.
Hi, is the model only support ref2va or it can use in fl2va?
the workflow v18 looks ref2va and fl2va has different model?
Supports all.
@Darksidewalker ok~ thank!!
I test it, it's good for anime ref2va.
I got color shift and contrast change . and sometimes it have some ghosting box in the videos .
Hope @Darksidewalker will fix that in the future <3 .
Btw , i love the motion so much , it is perfect for character animation <3
fixed
@Darksidewalker Thank you so much , brother ! <3
I'm trying to use the 4 or 8 turbo checkpoints to preview, but I'm finding that it does not really follow the prompt well compared to the base version. Is there any other lora's that follows the prompt well.
Wow, this model and your workflow nodes are a game-changer. Seriously impressive work. Non-turbo works really really well. Turbo has given me some anatomy issues. Overall really great model, LTX is left in the dirt
Thanks for the update! Is Turbo v2 recommended at 4 steps or 8?
Obviously 8-steps are more quality. Turbo is always at a cost.