VDN-H3 Turbo Goes Standalone: 8-Step LoRA for MiniMax H3
drbaph extracted the VDN-H3 Turbo adapter into a standalone 8-step LoRA for MiniMax H3, with pruned fl2va/ref2va variants plus an INT8 ConvRot VDN checkpoint for Comfy Kitchen.
Why standalone matters
Until now the only way to run the VDN-H3 Turbo adapter was through the ComfyUI-VDN-H3 custom nodes with the full hybrid-attention patch: the 8-step checkpoint, its linear branch, and both adapters loaded together. The new standalone LoRA strips the Turbo adapter out of that stack and repackages it as a plain MiniMax H3 LoRA, so it loads with ComfyUI's built-in LoRA loader on the stock H3 checkpoint, no custom node required.
Per drbaph, the Turbo adapter inside VDN-H3 was initialized from larryvrh's MiniMax H3 Turbo LoRA and then further DMD-trained as part of VDN-H3. This release is the standalone conversion of that resulting adapter, so it carries distillation signal from both lineages.
What was released
The LoRAs live in the experimental/ folder of drbaph/MiniMax-H3-Turbo-Lora-ComfyUI:
| File | Base | Notes |
|---|---|---|
minimax_h3_dmd_8step_turbo.safetensors | non-pruned fl2va / ref2va | one LoRA for both unpruned variants |
minimax_h3_dmd_fl2va_8step_turbo_pruned.safetensors | pruned fl2va | separate variant for the pruned/curve-form checkpoint |
minimax_h3_dmd_ref2va_8step_turbo_pruned.safetensors | pruned ref2va | separate variant for the pruned/curve-form checkpoint |
Recommended settings: 8 steps, strength 0.65 to 1.0. Each pruned file is about 2.1 GB.
The INT8 ConvRot VDN checkpoint
The second release targets the full VDN-H3 route. drbaph/vdn-minimax-h3-int8-convrot-comfyui pre-quantizes the 8-step stage-dmd-step-250 checkpoint into Comfy Kitchen INT8 ConvRot format, the same serialization ComfyUI ships for its own int8-convrot diffusion models. Only the linear branch's dispatched matmul weights are quantized; adapters, biases, norms and convolutions pass through untouched.
| bf16 original | INT8 ConvRot | |
|---|---|---|
| linear branch | 4.28 GB | 2.20 GB |
| stage total | 5.09 GB | 3.25 GB |
| branch matmuls | 1x | ~2.7x faster |
| end-to-end (single test run) | ~111 s | ~95 s |
An A/B render at 1280x736 / 61 frames with the same seed was visually identical, and peak VRAM while loading dropped by roughly 4.7 GB. The quantized stage requires ComfyUI-VDN-H3 v1.3.0 or newer and drops into ComfyUI/models/vdn/ like the original.
Ref2VA example rendered with the INT8 ConvRot stage: 8 steps, er_sde / beta, 928x928.
How it fits with the other H3 accelerators
Community testing in the Banodoco H3 channels this week put the numbers in context: the VDN route trades some speed against the lightx2v turbo LoRAs (one benchmark timed the 8-step VDN Turbo at 2:04 for 1280x736 versus 1:24 for the 4-step lightx2v turbo on the same hardware) but holds fast motion better than the FastH3 route. The standalone 8-step LoRA now gives a middle option: VDN's DMD-distilled adapter without loading the linear branch at all.
For 50-step audio work, drbaph's suggested recipe is to generate at 512x512 minimum and upscale to 1080p through the ref2va turbo 8-step path, downscaling to 256 first for speed.
Availability
- Standalone Turbo LoRA: drbaph/MiniMax-H3-Turbo-Lora-ComfyUI,
experimental/folder (MiniMax H3 Community License via the base model; the repo itself is Apache-2.0) - INT8 ConvRot VDN stage: drbaph/vdn-minimax-h3-int8-convrot-comfyui (MiniMax H3 Community License)
- Custom nodes for the full VDN route: Saganaki22/ComfyUI-VDN-H3 v1.3.0+
Comments
Sign in with GitHub to join the discussion.