<div data-youtube-video="https://youtu.be/g3QmFCf2dF4"><iframe src="https://www.youtube.com/embed/g3QmFCf2dF4" width="853" height="480" allowfullscreen></iframe></div> 
 
RunningHub workflow page: https://www.runninghub.ai/zh-cn/post/2098328697479983105?inviteCode=rh-v1111Bilibili tutorial video: https://www.bilibili.com/video/BV16DYL6FEsh/# PDD — Endpoint Wiring with Task-Mode Review## Purpose and preparationResolve 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.Image 5 connects first_frame and 30 last_frame. The base/setup select FL2VA but the conditioner task says I2VA first-frame. Check the matching implementation before promising endpoint control; no silent repair or demonstrated failure is asserted.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-FL2VA-Acc-8Step_comfyui_pdd-T8.safetensors`- `minimax_h3_video_vae_int8_convrot.safetensors`- `qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors`- `minimax_h3_audio_vae_fp32.safetensors`- `minimax_h3_fl2va_int8_convrot.safetensors`## Version and rightsNew 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/2098317966491340801/?inviteCode=rh-v1111)Links not live-verified. The workflow is supplied through AI-KSK.

MiniMax H3 PDD 8-step First-Last Frame Video

<div data-youtube-video="https://youtu.be/g3QmFCf2dF4"></div> 
 
RunningHub workflow page: https://www.runninghub.ai/zh-cn/post/2098328697479983105?inviteCode=rh-v1111Bilibili tutorial video: https://www.bilibili.com/video/BV16DYL6FEsh/# PDD — Endpoint Wiring with Task-Mode Review## Purpose and preparationResolve 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.Image 5 connects first_frame and 30 last_frame. The base/setup select FL2VA but the conditioner task says I2VA first-frame. Check the matching implementation before promising endpoint control; no silent repair or demonstrated failure is asserted.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-FL2VA-Acc-8Step_comfyui_pdd-T8.safetensors`- `minimax_h3_video_vae_int8_convrot.safetensors`- `qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors`- `minimax_h3_audio_vae_fp32.safetensors`- `minimax_h3_fl2va_int8_convrot.safetensors`## Version and rightsNew 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/2098317966491340801/?inviteCode=rh-v1111)Links not live-verified. The workflow is supplied through AI-KSK.

Files with SHA256: aa477ff6e4c90f805e6d1efefbaf1dd150f0adb50f4842c78f94ee790e299f38