MiniMax-H3-Acc-LoRAs

MiniMax-H3-Acc-LoRAs

Introduction

We apply Parallel Decoding Distillation (PDD) ¹ to MiniMax-H3, enabling efficient video generation in only a few inference steps.

For more details, please refer to our GitHub repo.

Demo

FL2VA (768p)

<table border="0" style="width: 100%; text-align: center; margin-top: 20px;"> <thead> <tr> <th style="text-align: center;" width="33%">MiniMax-H3-FL2VA</th> <th style="text-align: center;" width="33%"><a href="https://huggingface.co/lightx2v/Minimax-h3-Turbo/blob/main/minimax_h3_fl2v_turbo_4step_v1.1_768p_bf16.safetensors">Minimax-h3-Turbo <br> (fl2v_turbo_4step_v1.1_768p)</a></th> <th style="text-align: center;" width="33%">MiniMax-H3-FL2VA-Acc-8Step </th> </tr> </thead> <tr> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_baseline_1.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_turbo_4step_v1.1_768p_1.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_acc_8step_v1_1.mp4" width="100%" controls loop></video> </td> </tr> <tr> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_baseline_2.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_turbo_4step_v1.1_768p_2.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_acc_8step_v1_2.mp4" width="100%" controls loop></video> </td> </tr> <tr> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_baseline_3.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_turbo_4step_v1.1_768p_3.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_fl2va_acc_8step_v1_3.mp4" width="100%" controls loop></video> </td> </tr> </table>

Ref2VA

<table border="0" style="width: 100%; text-align: center; margin-top: 20px;"> <thead> <tr> <th style="text-align: center;" width="33%">MiniMax-H3-Ref2VA</th> <th style="text-align: center;" width="33%"><a href="https://huggingface.co/lightx2v/Minimax-h3-Turbo/blob/main/minimax_h3_ref2v_turbo_4step_v0.1_bf16.safetensors">Minimax-h3-Turbo <br> (ref2v_turbo_4step_v0.1)</a></th> <th style="text-align: center;" width="33%">MiniMax-H3-Ref2VA-Acc-8Step</th> </tr> </thead> <tr> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_baseline_1.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_turbo_4step_v0.1_768p_1.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_acc_8step_v1_1.mp4" width="100%" controls loop></video> </td> </tr> <tr> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_baseline_2.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_turbo_4step_v0.1_768p_2.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_acc_8step_v1_2.mp4" width="100%" controls loop></video> </td> </tr> <tr> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_baseline_3.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_turbo_4step_v0.1_768p_3.mp4" width="100%" controls loop></video> </td> <td> <video src="https://huggingface.co/alibaba-pai/MiniMax-H3-Acc-LoRAs/resolve/main/results/minimax_h3_ref2va_acc_8step_v1_3.mp4" width="100%" controls loop></video> </td> </tr> </table>

[!NOTE] The above test cases are from <a href="https://github.com/ModelTC/Minimax-H3-Turbo/tree/main/examples">Minimax-H3-Turbo</a>. Videos are generated with a LoRA weight of 1.0 at both 4 and 8 NFE.

Quick Start

Set model_path and pdd_lora_path to the MiniMax-H3 model and the matching acceleration LoRA checkpoint in predict_t2v.py for FL2VA or predict_ref2v.py for Ref2VA, then run the corresponding script. Each example uses apply_pdd_lora to load the checkpoint and derive the required number of inference steps from its configuration.

[!NOTE] These scripts use Diffusers' MiniMax-H3 ModularPipeline and require diffusers >= 0.40.0.

Reference

<ol> <li id="ref1">Neta Shaul, et al. "Parallel Decoding Distillation for Fast Image and Video Generation.". arXiv preprint arXiv:2607.26004 (2026).</li> </ol>