sandpies/Minimax-H3-fl2va-ref2va-hybrid-unpruned-int8
52.3k
MiniMax H3 Hybrid (unpruned INT8) — FL2VA base + Ref2VA AdaLN
Community weight-selection merge of the unpruned MiniMax H3 int8_convrot checkpoints from Comfy-Org/MiniMax-H3.
Same hybrid idea as smhfacct/Minimax-H3-fl2va-ref2va-hybrid-models, but built from the full / unpruned INT8 pair (time_embedder) instead of the pruned pair (adaln_t_table).
What’s in these files
No training. For each tensor:
- Base:
minimax_h3_fl2va_int8_convrot.safetensors(unpruned) - Overlay:
blocks.{N}.adaln_proj.linear.{weight,bias}(and matching.comfy_quant/_scalesiblings) taken fromminimax_h3_ref2va_int8_convrot.safetensors(unpruned) for a late block range - Everything else stays FL2VA
Final-layer AdaLN and output heads remain FL2VA in both.
Pruned vs unpruned
Do not mix pruned and unpruned files in one load.
Use in ComfyUI
- Put a file in
ComfyUI/models/diffusion_models/ - Load with stock Load Diffusion Model /
UNETLoader - Treat the model as Ref2VA for reference packing (
MiniMaxH3ReferenceToVideo, HandTie, etc.)
Optional: the same merge can be done at runtime with scottmudge/ComfyUI_MinimaxH3HybridLoader (block_range_adaln).
Intended use
Reference-conditioned audio+video generation where you want FL2VA-like fidelity with Ref2VA’s late-block AdaLN reference routing, on the unpruned INT8 stack.
Limitations
- Experimental / subjective block ranges; try b30–49 first if you care more about raw quality, b25–49 if reference adherence feels weak.
- Inherits MiniMax H3 licensing and usage terms from the source checkpoints. This merge grants no extra rights.
Acknowledgements
- MiniMax for MiniMax H3
- Comfy-Org for the published INT8 convrot weights
- smhfacct for documenting / shipping the pruned hybrid approach this mirrors
