MiniMax H3 PDD 8-step Video-to-Video with Reference

RunningHub workflow page: https://www.runninghub.ai/zh-cn/post/2098329200986390529?inviteCode=rh-v1111

Bilibili tutorial video: https://www.bilibili.com/video/BV16DYL6FEsh/

# PDD — Reference-Image Generation

## Purpose and preparation

Resolve the exact model and node versions, inspect task/input agreement and choose authorized images before queueing. Treat saved controls as a starting point, not accepted output evidence.

One image 5 connects ref_image_0. Ref2VA base, PDD setup variant and conditioner task are selected.

The PDD setup receives the base model and AV latent and supplies the model, sampler and sigmas. Strength is 1; no numeric sigma array is serialized and report_json is unused. The eight-step class/asset name identifies a route, not measured NFE or wall-clock performance.

Conditioning stores 736×416, length 124 without size/time links, with native audio. Noise control randomizes. Sampler OUTPUT reaches AVDecode, then VHS frames/generated_audio. VHS stores 24fps/H.264/CRF19/yuv420p/trim_to_audio=false. Its AnimateDiff prefix is not architecture evidence.

At24fps, 124 output frames would be nominally about5.167 seconds, not a measured duration. All enabled soundtracks come from decoded generated audio; source audio inputs are absent and conditioner mux_audio is unused. Prompt soundscapes are instructions, not audible proof. Inspect actual task resolution, images, timing, speech and endpoints in a small authorized-media test.

## Files and selected assets

- [Workflow JSON]()

- [Selected cover]()

- `MiniMax-H3-Ref2VA-Acc-8Step_comfyui_pdd-T8.safetensors`

- `minimax_h3_video_vae_int8_convrot.safetensors`

- `minimax_h3_audio_vae_fp32.safetensors`

- `qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors`

- `minimax_h3_ref2va_int8_convrot.safetensors`

## Version and rights

New copies retain the original configuration except recorded historical preview cleanup. No task-mode mismatch is silently repaired. This is descriptive local copy, not a published upgrade.

No inference, visual/audio acceptance, hardware timing or license verification was performed. Covers are selected same-stem WebP assets, not accepted output samples. Weights and external input media are not bundled. Check workflow, nodes, models, LoRAs, images and cover permissions separately; no blanket commercial license or speed winner is asserted.

[Source-provided tutorial](https://www.youtube.com/watch?v=g3QmFCf2dF4) · [Source workflow entry](https://www.runninghub.cn/post/2098326349755805698/?inviteCode=rh-v1111)

Links not live-verified. The workflow is supplied through AI-KSK.