MiniMax-H3-Pruned-Ref-Delta-Fused-r1024-ComfyUI

About

What MiniMax-H3-Pruned-Ref-Delta-Fused-r1024-ComfyUI is

MiniMax-H3 Pruned Ref-Delta Fused r1024 — ComfyUI Single File Native ComfyUI-format single-file conversion of diffusers-modular/MiniMax-H3-Pruned-Ref-Delta-Fused-r1024. This repository contains the MiniMax-H3 diffusion transformer only. It does not include the text encoder, tokenizer, VAE, or the rest of the MiniMax-H3 pipeline. The BF16 checkpoint is a state-dict/layout conversion of the immediate source checkpoint. Four native-ComfyUI INT8 derivatives are provided: full core-Linear INT8 and INT8 ConvRot variants, plus fc2-BF16 compatibility variants retained for older/problematic ComfyUI execution paths. No training, fine-tuning, additional pruning, or learned-weight adaptation was performed. Available checkpoints The ~21.0 GB sizes above are the rounded sizes reported by hf/Xet during upload. SHA-256 checksums BF16 full-file SHA-256: BF16 tensor-data-region SHA-256: Full INT8 full-file SHA-256 values: Compatibility-variant full-file SHA-256 values: INT8 quantization policy All four quantized checkpoints use ComfyUI's native per-layer .comfyquant format and TensorWiseINT8Layout. No custom quantized-model loader is required. The following weights are quantized in every one of the 50 main transformer blocks in all INT8 variants: That accounts for 150 quantized core Linear layers. The full INT8 variants additionally quantize: That adds another 50 layers for 200 quantized core Linear layers total. The -fc2bf16.safetensors compatibility variants keep those 50 fc2 weights in BF16 and therefore retain the original 150-layer INT8 policy. All smaller/sensitive tensors remain in their source precision, including the pruned AdaLN table and projections, final-layer projections, norms, patch/text projections, and token refiner. Regular INT8 uses tensor-wise scaling: INT8 ConvRot uses: The two Diffusers-only auxiliaries adalnbasis and adalnmean, which are retained in the repaired BF16 artifact but unused by native ComfyUI inference, are omitted from all quantized derivatives. fc2 compatibility history The first full 200-layer INT8 test also quantized blocks.N.mlp.fc2.weight. In the ComfyUI environment used for the initial conversion work, large MiniMax-H3 sequences could fa

Summarised from the published model card. Read the full card on the HuggingFace links below.

Specifications

What it is

How it works

How video models work

Variants

Sizes & precisions

Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys minimax-h3-pruned-ref-delta-fused-r1024-comfyui for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (minimax-h3-pruned-ref-delta-fused-r1024-comfyui below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/videos/generations \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"minimax-h3-pruned-ref-delta-fused-r1024-comfyui","prompt":"a drone shot over a forest"}'

Details

Languages, data & research

Licence

Commercial licence needed

Sources

Weights & code

Explore

More video models