--- title: vdn-minimax-h3-int8-convrot-comfyui canonical_url: "https://www.modelscope.cn/models/drbaph/vdn-minimax-h3-int8-convrot-comfyui" md_url: "https://www.modelscope.cn/models/drbaph/vdn-minimax-h3-int8-convrot-comfyui.md" repository: drbaph/vdn-minimax-h3-int8-convrot-comfyui last_updated: 2026-09-14 license: other parameters: 2.7B tensor_type: - I8 - F32 - BF16 - U8 library_name: - safetensors - pytorch frameworks: - pytorch downloads: 14493 stars: 1 tags: - comfyui - video - int8 - quantization - minimax-h3 - vdn --- # vdn-minimax-h3-int8-convrot-comfyui > vdn-minimax-h3-int8-convrot-comfyui - drbaph 在 ModelScope 开源的模型。VDN-H3 stage-dmd-step-250 — INT8 ConvRot (ComfyUI pre-quantized) drbaph/vdn-minimax-h3-int8-convrot-comfyui 是 ModelScope 魔搭社区上的 2.7B 参数机器学习模型，采用 other 许可。 - **Repository**: drbaph/vdn-minimax-h3-int8-convrot-comfyui - **License**: other - **Parameters**: 2.7B - **Tags**: comfyui, video, int8, quantization, minimax-h3, vdn - **Downloads**: 14493 - **Stars**: 1 - **Last updated**: 2026-09-14 Source: https://www.modelscope.cn/models/drbaph/vdn-minimax-h3-int8-convrot-comfyui --- # VDN-H3 `stage-dmd-step-250` — INT8 ConvRot (ComfyUI pre-quantized) Pre-quantized **Comfy Kitchen INT8 ConvRot** version of the 8-step checkpoint (`stage-dmd-step-250`) from [OpenVDN/vdn-minimax-h3](https://huggingface.co/OpenVDN/vdn-minimax-h3). For use with [ComfyUI-VDN-H3](https://github.com/Saganaki22/ComfyUI-VDN-H3).

> **Requires ComfyUI-VDN-H3 v1.3.0+** ([release](https://github.com/Saganaki22/ComfyUI-VDN-H3/releases/tag/v1.3.0)). Same weights, same adapters, A/B-verified visually identical output. What you save: | | original stage | this repo | |---|---|---| | linear branch | 4.28 GB bf16 | 2.20 GB int8 | | stage total | 5.09 GB | 3.25 GB | | branch VRAM (`cache_gpu`) | ~4.3 GB | ~2.2 GB | | peak VRAM while loading | — | ~4.7 GB lower | Faster: the branch's dispatched matmuls run **2.7x** faster than bf16, and the A/B render finished **~1.2x** faster end-to-end (~111 s -> ~95 s, single run each). ## BF16 vs INT8 ConvRot — A/B 8 steps, er_sde / beta, 1280x736 / 61 frames, merge, `cache_gpu`, same seed. Identical output; wall clock from a single run each (±5 s sampling). | stage | precision | wall time | peak VRAM (min free) | video | |---|---|---|---|---| | `stage-dmd-step-250` | bf16 | ~111 s | 3.6 GB free |

| ## Install Download this repo into `ComfyUI/models/vdn/` (the folder name becomes the `vdn_checkpoint` entry), then select it in the Apply-VDN node's `vdn_checkpoint` dropdown. Nothing else changes; the base MiniMax-H3 model and this node's settings work exactly as before. ## Ref2V example Rendered with this repo's INT8 ConvRot stage on a `ref2va` base: 8 steps, er_sde / beta, 928x928.

## How it was made `tools/quantize_vdn_branch_int8.py` from the ComfyUI-VDN-H3 repo, using comfy-kitchen's `TensorWiseINT8Layout` (convrot, per-channel) — the same serialization ComfyUI ships for its own int8-convrot diffusion models. Only the branch's dispatched linear weights are quantized; adapters, biases, norms and convs pass through untouched. ## Acknowledgement VDN-H3 is built on MiniMax-H3 and starts from its released transformer weights. We also thank Diffusers, FlashAttention, and Triton, which the optimized inference path is built with. ## License VDN-H3 is a derivative of MiniMax-H3 and is distributed under the MiniMax H3 Community License Agreement, included here verbatim from the upstream repository. The agreement grants rights only in its applicable territory, defined as worldwide excluding the European Union, the United Kingdom, the Republic of Korea, and the United States of America. It states that use outside the applicable territory is not authorized and invites people in an excluded territory to contact MiniMax about obtaining a license. The agreement also contains redistribution requirements and an Acceptable Use Policy. Among other requirements, a distribution must include the agreement, modified files must carry notices of their modification, and distributions to third parties other than through hosted services must include the NOTICE file supplied with the code repository. Please read the agreement in full before using or distributing VDN-H3. This note is not a substitute for the license text or for legal advice.