Text-to-Video
Diffusers
Safetensors
world-model
video-generation
image-to-video
interactive
distillation
Instructions to use AlayaLab/Evoke with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use AlayaLab/Evoke with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("AlayaLab/Evoke", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
| { | |
| "_comment": [ | |
| "This repository holds several EVOKE checkpoints. It is NOT a single loadable", | |
| "diffusers pipeline root -- there is no top-level vae/ or transformer/, and", | |
| "DiffusionPipeline.from_pretrained() on this repo id will not work.", | |
| "", | |
| "Every released model directory is the PARENT of a transformer/, and loads as", | |
| " from_pretrained(<model_dir>, subfolder=\"transformer\")", | |
| "with the shared components taken from evoke-base/.", | |
| "", | |
| "EVOKE also pins a development fork of diffusers that is not on PyPI. See", | |
| "https://github.com/AlayaLab/Evoke for installation, weights layout and inference." | |
| ], | |
| "_shipped_model": "evoke/stage3_post_distillation", | |
| "_base_components": "evoke-base", | |
| "_diffusers_fork_version": "0.37.0.dev0", | |
| "_models": { | |
| "evoke/stage1_camera_control": "multi-step camera-controllable model -- 50 steps, CFG 5.0", | |
| "evoke/stage2_few_step_training": "few-step distillation -- 3-step pyramid", | |
| "evoke/stage3_long_distillation": "30 s long-video distillation -- post-distill init", | |
| "evoke/stage3_post_distillation": "the shipped model -- 3 steps, CFG-free", | |
| "evoke/evoke_teacher": "dual-expert DMD teacher, {high,low}_noise -- training only" | |
| }, | |
| "_components": { | |
| "evoke/<model>/transformer": [ | |
| "evoke.modules.transformer_evoke", | |
| "EvokeTransformer3DModel" | |
| ], | |
| "evoke-base/vae": [ | |
| "diffusers", | |
| "AutoencoderKLWan" | |
| ], | |
| "evoke-base/text_encoder": [ | |
| "transformers", | |
| "UMT5EncoderModel" | |
| ], | |
| "evoke-base/tokenizer": [ | |
| "transformers", | |
| "T5TokenizerFast" | |
| ], | |
| "evoke-base/scheduler": [ | |
| "evoke.diffusers_version.scheduling_evoke_diffusers", | |
| "EvokeScheduler" | |
| ] | |
| }, | |
| "_external_weights": { | |
| "ViGeo": "pkqbajng/ViGeo -- REQUIRED depth backend, downloaded separately", | |
| "Depth-Anything-3": "depth-anything/da3-giant -- OPTIONAL, only for DEPTH_BACKEND=da3" | |
| } | |
| } | |