998

模型介绍

本模型依托魔搭社区（ModelScope）AIGC专区模型训练环境与算力完成训练。

• 模型类型：LoRA
• 基础模型：MiniMax/MiniMax-H3
• 训练代码：DiffSynth-Studio
• 训练数据量：30
• 总训练步数：1900
• 开源协议：Apache-2.0

推理代码

安装 DiffSynth-Studio：

pip install diffsynth

开始推理：

import torch
from diffsynth.pipelines.minimax_h3_audio_video import MiniMaxH3Pipeline, ModelConfig
from diffsynth.utils.data.audio_video import write_video_audio

vram_config = {
    "offload_dtype": "disk",
    "offload_device": "disk",
    "onload_dtype": "disk",
    "onload_device": "disk",
    "preparing_dtype": torch.bfloat16,
    "preparing_device": "cuda",
    "computation_dtype": torch.bfloat16,
    "computation_device": "cuda",
}
pipe = MiniMaxH3Pipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="MiniMax/MiniMax-H3", origin_file_pattern="FL2VA/text_encoder/model*.safetensors", **vram_config),
        ModelConfig(model_id="MiniMax/MiniMax-H3", origin_file_pattern="FL2VA/transformer/model*.safetensors", **vram_config),
        ModelConfig(model_id="MiniMax/MiniMax-H3", origin_file_pattern="FL2VA/video_vae/source/model.safetensors", **vram_config),
        ModelConfig(model_id="MiniMax/MiniMax-H3", origin_file_pattern="FL2VA/audio_vae/model.safetensors", **vram_config),
    ],
    processor_config=ModelConfig(model_id="MiniMax/MiniMax-H3", origin_file_pattern="FL2VA/processor/"),
    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 2,
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="yy3yy3/998", origin_file_pattern="998_c1-st1900.safetensors"))

prompt = "A cat"
video, audio = pipe(
    prompt=prompt,
    height=480, width=832, num_frames=124, num_inference_steps=50, seed=0,
)
write_video_audio(
    video=video, audio=audio,
    output_path="t2va.mp4", fps=24, audio_sample_rate=32000,
)