MiniMax H3 SparseRef15 Hybrid – FL2VA × Ref2VA INT8 ConvRot
A custom MiniMax H3 FL2VA / Ref2VA hybrid designed to balance FL2VA image quality and motion freedom with lightweight Ref2VA reference consistency.
Unlike conventional H3 hybrid models that replace one continuous range of later transformer blocks with Ref2VA blocks, SparseRef15 distributes Ref2VA AdaLN layers sparsely across the DiT.
Base:
minimax_h3_fl2va_pruned_int8_convrot
Reference overlay:
minimax_h3_ref2va_pruned_int8_convrot
Ref2VA AdaLN blocks:
1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31, 34, 37, 40, 43
Only 15 of the 50 transformer blocks use Ref2VA AdaLN conditioning.
The remaining blocks, including final_layer.adaln_proj, remain FL2VA-based.
The idea is to distribute relatively light Ref2VA influence across early, middle, and later parts of the network instead of concentrating it in a single block range.
Intended characteristics:
- Better reference retention than pure FL2VA
- More motion and prompt freedom than strongly Ref2VA-weighted hybrids
- Reduced tendency toward overly rigid reference following
- Good balance between character/reference consistency and FL2VA visual quality
- Especially interesting for chained or multi-stage video workflows
IMPORTANT:
No Lightning / Turbo / acceleration LoRA is merged into this checkpoint.
You can freely use your preferred MiniMax H3 acceleration LoRA separately.
The use of ModelSamplingMiniMaxH3 is not recommended.
Recommended acceleration LoRA:
MiniMax H3 FL2V LightX2V Turbo 4-step v0.1
LightX2V v0.1 is a good starting point when stability, image quality, and chained video consistency are more important than maximum motion intensity.
Other acceleration LoRAs, including DARE-TIES based merges or stronger Turbo LoRAs, may also work well if more aggressive motion is desired.
Recommended starting points:
Stable / chained video:
- LightX2V Turbo 4-step v0.1
- Euler
- simple scheduler
- Video Shift around 12
- Ref video around 12–24 frames, adjusted depending on desired reference strength
Higher visual impact / more dynamic results:
- ER SDE
- beta scheduler
- Video Shift around 16~28
ER SDE + beta is also highly recommended and can produce particularly clean and visually rich results.
For long chained workflows, Euler + simple may still be the safer starting point when maximum continuity is the priority.
The model retains the original pruned INT8 ConvRot format.
No additional FP8 conversion or re-quantization was applied.
This is an experimental custom hybrid, not an officially trained MiniMax model.
Results will vary depending on prompt, reference images/videos, reference length, sampler, scheduler, Video Shift, acceleration LoRA, and workflow design.
Description
FAQ
Comments (37)
Did you actually use the int8 convrot models for the merge?
Yes. I used minimax_h3_fl2va_pruned_int8_convrot as the base and minimax_h3_ref2va_pruned_int8_convrot as the overlay. The MiniMax H3 Hybrid Loader is quantization-aware and preserves the INT8 ConvRot / .comfy_quant structure when overlaying the selected AdaLN blocks.
Thanks for trying SparseRef15.
This is an experimental distributed FL2VA / Ref2VA hybrid.
I’d love to hear what works best for you: short clips, chained video, stronger reference use, or something else.
testing it now, hope i'll get good results, uploading soon some pdd lora tests vs dareties lora with the hybrid checkpoint 30-49. would like really see the diff with your checkpoint
@vAnN47
Personally, I recommend LightX2V Turbo 4-step v0.1.
Your model is not NSFW. Your examples are incorrect🫤
Should I not have used it to mean that it is capable of generating NSFW content?
For this one, I think your prompting skills might not be up to par.
@g1263495582
This is not a model exclusively for NSFW content. I apologize if I caused any misunderstanding.
Uh, MiniMaxH3 by default is NSFW. No need to apologize @Aki7777777 because it's obvious this person is trying to sow discord.
Good job! But very Asian oriented ;) all my caucasian girls end asian
Thanks! Though, I haven't really made any changes to the default settings...
I tried it on non-Asians, too. What do you think?
@Aki7777777 wow looking real good... makes me wonder why my references end up looking asian hahaha good job :)
It's not your imagination: it definitely converted my caucasian subjects to Asian, especially once they were far enough from the camera.
@mwoody450 @kingdan78
Interesting. Since this hybrid is essentially based on minimax_h3_ref2va_pruned_int8_convrot, the tendency may originate from the original Ref2VA model. I haven't seen it in my own generations, so it may be more noticeable with extremely long single-clip generations.
Regardless of quality if model can draw only asians its auto skip.
I have simply combined off-the-shelf models; no intentional customization has been performed.
Did it ever occur to you that you can prompt for other races? You should try it.
It is a very excellent model. Thank you.
Well, this replaced the regular ref2va for me! It's not even close, I much prefer this one. Movement is more natural. Physics make more sense. Prompts are better followed. It seems to let me use ref images for styling, that don't immediately become part of the events. Very nice!
So far, I've just used it like the regular ref2va model. I do prefer the 4step_v1.1_768p turbo lora with it.
Thank you for your efforts!
Thank you so much for the detailed feedback!
This is very close to what I hoped the sparse hybrid structure might achieve — keeping useful Ref2VA conditioning while giving FL2VA more freedom for motion and prompt interpretation.
Your note about using reference images mainly for styling without having them immediately become part of the events is especially interesting. I'll definitely keep an eye on whether other users observe the same behavior.
Also, thanks for the 4step_v1.1_768p Turbo LoRA recommendation. I haven't tested that combination enough yet, so I'll give it a try!
@Aki7777777 the new turbo lora v1.1 4 step 768 is for FL2VA, not R2V, and it not work good for R2V..
If you want production ready quality for your R2V video, just use the turbo v4 step600 ema hybrid lora, that work for both model. with at least 8 steps.
@Pat3dx
Thank you for the advice.
As stated in the description, this model is based on FL2VA, so it is perfectly fine to use the FL2VA Fast LoRA with it.
@Aki7777777 Yes, in the short term, such as 5S, it may be executed immediately, but in the long term, such as 10s, there will be a process from the beginning to the reference posture, but this can be controlled through prompt words.
The characters set will reference the poses in the pose map in the specified environment, instead of completely replicating the pose map like the original version.
@big27916430
Thanks a lot for the detailed feedback!
I'm really glad to hear the timing-based prompt control is working well for you, especially for transitions toward the reference pose in longer clips.
It’s also great to hear that the imitative action behavior is stronger than in the original version.
I really appreciate your testing and support!
I see the fine detail quality and video consistency is much better than with pure ref2va model. unfortunately reference following is noticeable worse than pure ref2va model. so I would wish for a version that tries to keep more ref following intact, even maybe at small cost to quality.
Thanks for the feedback!
I'm actually working on a version that keeps stronger Ref2VA reference following, even if it comes at a small cost to detail quality or consistency. I'm still testing the balance, but I'll share it once it's ready.
@Aki7777777 I will wait for it!
That's really cool. It's fascinating. Does this model represent the genital area better than the stock base model?
Thank you for your feedback.
Essentially, it is based on FL2v, with Ref2VA AdaLN layers distributed and referenced at intervals throughout the DiT.
If you are looking for even more realistic rendering or movement, I recommend using FL2v-based LoRAs.
been testing yours and eros3, yours is better overall when using FL2AV for ref workflow, the trade-off is to sacrifiy the first frame which nothing to me, Thank you.
Just minor suggestion, may you check this technique out? https://huggingface.co/cicalooo/10Eros-Max-h3-int8-convrot/blob/main/SKIP_EDGES.md they claim the the first and last transformer blocks would be fully taken in account.
Thanks! I'm glad to hear it's working well for your FL2AV reference workflow.
I checked the skip-edges technique. It looks like it keeps blocks 0, 1, 48 and 49 in BF16 while quantizing the rest, rather than directly changing the reference weighting.
My current hybrid is built from already-quantized INT8 ConvRot checkpoints, so applying this properly would require rebuilding it from the BF16 source weights.
It's an interesting idea though, so I'll keep it in mind for a future test.
Testing complete. This is undoubtedly the best ref2va model I've tested! It features excellent prompt adherence and reference consistency, with outstanding image clarity.
One observation from my tests: applying LoRAs to other models often leads to severe structural distortions (e.g., human anatomy or perspective issues). For future updates, I highly recommend keeping the base model clean and letting users apply LoRAs on their own, rather than merging them by default.
Really appreciate your hard work and contribution to the community!
(P.S. Translated from AI, so please excuse any slight translation artifacts.)
Thanks for the detailed feedback!
No worries — this model does not have any LoRAs merged into it. It is built only by combining selected weights from the original FL2VA and Ref2VA models.
I also prefer to keep LoRAs separate so users can apply them freely depending on their workflow.
Would it be possible to make a bf16 version also?
BF16 version is definitely possible. I'm currently looking into higher-precision variants, although the file size will be quite large.