Image-to-Video
GGUF
MiniMax H3
Pika H3 Flex
rocmfpx
rocmfp4
amd
rocm
strix-halo
stable-diffusion.cpp
video-generation
audio-video-generation
quantized
Instructions to use Harsha001/Pika-H3-Flex-ROCmFPX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Pika H3 Flex
How to use Harsha001/Pika-H3-Flex-ROCmFPX with Pika H3 Flex:
# pip install -r examples/requirements-diffusers.txt (see the examples/ folder of this repo) import torch from diffusers import ComponentsManager, ModularPipeline from diffusers.utils.export_utils import encode_video from PIL import Image workflow = "fl2va" # "fl2va" (first-frame to video+audio) or "ref2va" (reference images/video to video+audio) manager = ComponentsManager() manager.enable_auto_cpu_offload(device="cuda:0") pipe = ModularPipeline.from_pretrained("Harsha001/Pika-H3-Flex-ROCmFPX", workflow=workflow, components_manager=manager) pipe.load_components( pretrained_model_name_or_path="Harsha001/Pika-H3-Flex-ROCmFPX", dtype={"vae": torch.float32, "audio_vae": torch.float32, "default": torch.bfloat16}, ) # The bundled base components are not Pika-adapted weights: load the matching adapter. pipe.load_lora_weights("Harsha001/Pika-H3-Flex-ROCmFPX", weight_name=f"{workflow}/adapter.safetensors") result = pipe( prompt="A cat walks through a sunlit cafe and jumps onto the counter", image=Image.open("first_frame.png").convert("RGB"), height=480, width=832, num_frames=121, generator=torch.Generator("cuda:0").manual_seed(0), output=["videos", "audio", "sampling_rate"], ) encode_video( result["videos"][0], fps=24, output_path="output.mp4", audio=result["audio"][0].float().cpu(), audio_sample_rate=result["sampling_rate"], ) - Notebooks
- Google Colab
- Kaggle
Confirm your location and accept the MiniMax H3 Community License
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
The MiniMax H3 Community License Agreement grants rights only outside the Excluded Territories: the European Union, the United Kingdom, the Republic of Korea and the United States of America. Use, reproduction, distribution and outputs inside those territories are not licensed.
Log in or Sign Up to review the conditions and access this model content.
Gated model You can list files but not access them
Preview of files found in this repository