Text-to-Video
Diffusers
Safetensors
English
MotifVideoPipeline
image-to-video
video-generation
diffusion-transformer
Instructions to use Motif-Technologies/Motif-Video-2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Motif-Technologies/Motif-Video-2B with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Motif-Technologies/Motif-Video-2B", dtype=torch.bfloat16, device_map="cuda") prompt = "A vibrant blue jay perches gracefully on a slender branch, its feathers shimmering in the soft morning light. The bird's keen eyes scan the surroundings, capturing the essence of the tranquil forest. It flutters its wings briefly, showcasing the intricate patterns of blue, white, and black on its plumage. The background reveals a lush canopy of green leaves, with rays of sunlight filtering through, creating a dappled effect on the forest floor. The blue jay then tilts its head, emitting a melodious call that echoes through the serene woodland, adding a touch of magic to the peaceful scene." image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
feat: GGUF + SageAttention guide β DPMSolver++ default, benchmark, README update
#13
by gkalstn0 - opened
Remove custom pipeline files (assumes diffusers PR #13551 merge). Rewrite inference.py with native MotifVideoPipeline, DPMSolver++ default, SageAttention option. README updated with GGUF + SageAttention section, benchmark table (9 variants x SDPA/Sage, ~1.59x speedup on H200), visual comparison webp.
Note: Written against current PR #13551 state (pre-merge). Includes workarounds (FlowDPMSolver subclass, explicit negative_prompt/LQ=False/frame_rate=24, revision=diffusers-integration). Will be simplified after PR merge.
gkalstn0 changed pull request status to closed