aptech0081/MiniMax-H3-Acc-LoRAs-ComfyUI
text-to-video model by aptech0081
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Earlier source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
ComfyUI-key repackaging of the official alibaba-pai/MiniMax-H3-Acc-LoRAs 8-step PDD acceleration LoRAs for MiniMax-H3 — full audio+video generation in 8 (or 4) sampler steps, CFG-free. These are not plain LoRAs. Each file carries a rank-64 trunk LoRA plus a Parallel Decoding Distillation head bank (32 per-interval final-layer projections per modality, PDD — arXiv:2607.26004). Loading them requires the companion custom node pack: ➡️ Jalen-Brunson/ComfyUI-MiniMax-H3-PDD-Acc (also loads the original alibaba-pai files directly — this repo just saves you the in-memory conversion and gives you inspectable standard LoRA keys). Put the LoRA files in ComfyUI/models/pddacc/. Pair FL2VA with an fl2va UNET, Ref2VA with ref2va (bf16 or int8-convrot builds both work). The baked checkpoint goes in ComfyUI/models/diffusionmodels/ instead. UNETLoader → MiniMaxH3SigmaShift (12/3) → MiniMax H3 PDD Acc LoRA (Apply) → BasicGuider (CFG 1.0), sampler euler, sigmas = the Apply node's sigmas output (the trained PDD block boundaries) into SamplerCustomAdvanced. Strengths 1.0, nfe 8 (4 is also…
The latest refresh did not succeed. This is the last available preview; check the original for changes.
Read the full model card ↗ · Preview checked 2026-10-20T02:17:37.331Z
Inside the original model card — Document outline
- MiniMax-H3 Acc LoRAs — ComfyUI conversion
- Files
- Usage (recipe is mandatory)
- Baked Ref2VA trunk (for cards that can't fully load the model)
- What was converted
- Credits
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. Reported project/model license: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
aptech0081/MiniMax-H3-Acc-LoRAs-ComfyUI is a text-to-video model repository published by aptech0081 on Hugging Face. The source reports the minimax-h3 library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- text-to-video
- Library
- minimax-h3
- Recent downloads (30 days)
- 27,589
- Cumulative likes
- 82
- Hugging Face trending score
- 1
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- apache-2.0
- Access
- Not gated by Hugging Face
- Created
- 2026-08-26T15:03:46.000Z
- Last modified
- 2026-08-30T22:14:04.000Z
Compatibility and lineage
Language coverage was not reported.
Reported base models: alibaba-pai/MiniMax-H3-Acc-LoRAs, MiniMaxAI/MiniMax-H3
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-10-19T06:44:15.629Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related models
QuantStack/Wan2.2-TI2V-5B-GGUF
QuantStack/Wan2.2-T2V-A14B-GGUF