| --- |
| license: apache-2.0 |
| pipeline_tag: text-to-video |
| inference: false |
| tags: |
| - world-model |
| - video-generation |
| - text-to-video |
| - image-to-video |
| - interactive |
| - distillation |
| --- |
| |
| <h1 align="center">Alaya-EVOKE: From Linear-Scaling Supervision to Endless World</h1> |
|
|
| <p align="center"> |
| <a href="https://github.com/AlayaLab/Evoke"><img src="https://img.shields.io/badge/GitHub-AlayaLab/Evoke-181717.svg?logo=github" alt="GitHub"></a> |
| <a href="https://evoke-world.github.io/Evoke/"><img src="https://img.shields.io/badge/🌐_Project_Page-evoke--world.github.io-1a73e8.svg" alt="Project Page"></a> |
| <a href="https://arxiv.org/abs/2608.13546"><img src="https://img.shields.io/badge/arXiv-2608.13546-b31b1b.svg" alt="arXiv"></a> |
| <a href="https://huggingface.co/papers/2608.13546"><img src="https://img.shields.io/badge/🤗_Paper-2608.13546-ffce1c.svg" alt="Paper page"></a> |
| </p> |
|
|
| Model weights for **EVOKE** ([paper](https://huggingface.co/papers/2608.13546)), a 3-step, CFG-free |
| interactive world model that generates **384 × 640 @ 24 fps** video and stays coherent over 30 s |
| rollouts. Code, docs and demos live in the GitHub repository — **this repository holds weights only.** |
|
|
| - ⚡ **3 steps, zero CFG** — 1.5 s of video every 2.11 s on one H200, one forward per step. |
| - 🌍 **Endless, not windowed** — scene geometry lives in an external camera-indexed world state bank, |
| so the denoiser context stays bounded however long the session runs. |
| - 🎛️ **Re-promptable mid-flight** — change the prompt while the rollout is running, no cut, no restart. |
|
|
| ## Contents |
|
|
| Every EVOKE directory is the **parent** of a `transformer/`, because it loads as |
| `from_pretrained(path, subfolder="transformer")`. |
|
|
| ``` |
| evoke-base/ vae / text_encoder / tokenizer / scheduler only |
| evoke/ |
| ├── stage1_camera_control/transformer/ multi-step camera-controllable model |
| ├── stage2_few_step_training/transformer/ few-step distillation (3-step pyramid) |
| ├── stage3_long_distillation/transformer/ 30 s long-video distillation (post-distill init) |
| ├── stage3_post_distillation/transformer/ the shipped model |
| └── evoke_teacher/{high,low}_noise/ the two DMD teacher experts -- training only |
| ``` |
|
|
| ## Usage |
|
|
| ```bash |
| git clone https://github.com/AlayaLab/Evoke && cd Evoke |
| pip install -r requirements.txt |
| |
| hf download AlayaLab/Evoke --local-dir models |
| hf download pkqbajng/ViGeo --local-dir models/ViGeo1.1 # REQUIRED depth backend |
| |
| MODE=t2v NUM_CHUNKS=20 bash scripts/inference/infer_post_distill.sh |
| ``` |
|
|
| `ViGeo` is a separate download and is **required** — every shipped recipe uses it as the depth backend |
| behind the world state bank. Depth-Anything-3 is optional. Both ship under CC-BY-NC-4.0, which is more |
| restrictive than this repository's Apache-2.0; check their licences before any commercial use. |
|
|
| Inference modes, the mode × model matrix, hour-scale rollouts and training are documented in the |
| GitHub repository. |
|
|
| ## Notes |
|
|
| The distilled models were trained on v2v conditioning alone, so `MODE=i2v|t2v` on them is **zero-shot**; |
| only `stage1_camera_control` has all three modes in distribution. |
|
|
| The vae / text encoder / tokenizer / scheduler in `evoke-base/` come from the released |
| [Helios](https://github.com/PKU-YuanGroup/Helios) base, which traces them to Wan. The EVOKE teacher is |
| built on [LingBot-World](https://github.com/robbyant/lingbot-world). |
|
|
| ## Citation |
|
|
| ```bibtex |
| @article{evoke2026, |
| title = {Alaya-EVOKE: From Linear-Scaling Supervision to Endless World}, |
| author = {Yin, Yuanyang and Wang, Gongxuan and Zhan, Yifan and |
| Li, Chuanhao and Zhang, Kaipeng and Zhao, Feng}, |
| journal = {arXiv preprint arXiv:2608.13546}, |
| year = {2026}, |
| } |
| ``` |
|
|
