Downloads · 30 days
0
kunalmiind/10Eros-Max
10Eros-Max is a image-text-to-video model from kunalmiind. Use it for the image-text-to-video task on the model card, and read the license before you ship it in a product. The card lists the license as other.
Now that minimax has clarified their license; This is, and was previously, meant to be an NSFW capable model that still has base model capability and functions. "Eros" is the greek god of sex, not sure why that didn't…
Downloads · 30 days
0
Access
Public
Updated Sep 4, 2026
Repo size
243 GB
Likes
0
Public
Click a slice to open those files.
.safetensors243 GB · 100%
From the Hugging Face model README
Now that minimax has clarified their license; This is, and was previously, meant to be an NSFW capable model that still has base model capability and functions. "Eros" is the greek god of sex, not sure why that didn't catch on or why there is confusion about what the point of this was. It started with literally extracting NSFW Wan2.2 model and Krea2 data to make those changes with grafts.
Int8 versions https://huggingface.co/cicalooo/10Eros-Max-h3-int8-convrot
For beta_4 use euler/simple 6-8 steps on all modes. SLA and Sparsity and shift can be used to tweak off of that.
beta_3 is somewhat corrupted and is technically only a t2v model. The turbo used harms referenced starts in i2v and reference mode. The turbo was fixed in v4, I made my own hybrid turbo sum ranked merge to use. For v3 TURBO model use multires/simple 6 steps, er_sde/simple 6 steps. No cache or spectrum if doing reference, they cause accuracy loss. Hybrid model is built on the delta1024 h3 merge.
Regarding the grafts:
Due to a ton of confusion I've let Claude compile a full MD on the grafting, including code and methodology as it was mostly used to create code, document, and manage the graft project while I tested, architected, and mixed. I was keeping that to myself since maybe it'd actually be nice to have something to myself but after a handful of ignorant comments I'm open sourcing everything from it except the scripts themselves. You can hand the md back to claude or any competent agent and graft in the same way or have them explain it until you understand. No more dumbassery allowed now.
This is an evolving project subject to future fixes in H3 training. Training is clearly problematic, and my branch will rely on Sulphur H3 tunes. But I didn't want to wait for all that so I took the data from older models, grafting it in where the model needs it, grafting it to attn layers at a low level that doesn't disturb H3's visual or audio output quality. That's essentialy it.
The standard H3 community license applies. Because this release now carries transferred character from LTX 2.3, Wan 2.2, and Krea 2, the community licenses for those source models apply as well to the portions of character that came from each.