Instructions to use lightx2v/Minimax-h3-Turbo-SLA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Inference
- Notebooks
- Google Colab
- Kaggle
MiniMax-H3 Turbo-SLA
MiniMax-H3 Turbo-SLA is a 4-step distilled FL2V checkpoint for MiniMax-H3, enhanced with SLA (SparseโLinear Attention) for more efficient inference.
The model uses an 85% attention sparsity ratio. In our tested LightX2V setup, it delivers approximately 2.5ร inference acceleration on an NVIDIA RTX 5090 while retaining competitive visual quality.
This repository contains LoRA weights and requires the original MiniMax-H3 model for inference. Actual performance may vary with resolution, video length, software environment, and hardware configuration.
Demo
Side-by-side comparison between the 30-step MiniMax-H3 base model and the 4-step MiniMax-H3 Turbo-SLA model:
Model Highlights
- 4-step distillation for substantially reduced denoising steps.
- SLA sparse attention with an 85% sparsity ratio.
- Approximately 2.5ร inference acceleration on RTX 5090 in our tested LightX2V setup.
- 768p FL2V generation support.
- Native LightX2V and ComfyUI-compatible BF16 checkpoints.
Available Checkpoints
| Checkpoint | Format | Description |
|---|---|---|
minimax_h3_fl2v_turbo_4step_v0.1_768p_sla_bf16.safetensors |
LightX2V | Native BF16 LoRA checkpoint for LightX2V inference. |
minimax_h3_fl2v_turbo_4step_v0.1_768p_sla_comfyui_bf16.safetensors |
ComfyUI | Converted BF16 LoRA checkpoint for ComfyUI workflows. |
LightX2V Inference
For installation and general MiniMax-H3 inference instructions, refer to the LightX2V MiniMax-H3 examples.
Use the following configuration for SLA-enabled inference on an RTX 5090:
minimax_h3_fp8_4step_5090_vae_fp8_sla.json
Key SLA-related settings in this configuration include:
{
"attn_type": "dynamic_sparse_attn",
"dynamic_sparse_attn_setting": {
"sparsity_ratio": 0.85,
"operator": "sage2"
},
"video_flow_shift": 6.0,
"audio_flow_shift": 3.0,
"h3_step_update": "training_euler"
}
Use the configuration file linked above as the source of truth for the complete inference setup, including FP8 DiT/VAE options and checkpoint paths.
ComfyUI
For ComfyUI workflows, use:
minimax_h3_fl2v_turbo_4step_v0.1_768p_sla_comfyui_bf16.safetensors
The ComfyUI checkpoint is a converted version of the SLA-enabled 4-step LoRA and is intended for compatible MiniMax-H3 ComfyUI workflows.
Related Projects
Acknowledgements
This work builds on MiniMax-H3, LightX2V, and SLA. We thank the respective authors and contributors for making their work available to the community.
Citation
If you use SLA in your work, please cite:
@article{zhang2025sla,
title={SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention},
author={Zhang, Jintao and Wang, Haoxu and Jiang, Kai and Yang, Shuo and Zheng, Kaiwen and Xi, Haocheng and Wang, Ziteng and Zhu, Hongzhou and Zhao, Min and Stoica, Ion and others},
journal={arXiv preprint arXiv:2509.24006},
year={2025}
}
License
The adapter weights in this repository are released under the Apache 2.0 License. Use of the MiniMax-H3 base model is also subject to its corresponding license and terms of use.
Model tree for lightx2v/Minimax-h3-Turbo-SLA
Base model
MiniMaxAI/MiniMax-H3