---
license: mit
library_name: open-oasis
datasets:
- microsoft/orca-agentinstruct-1M-v1
language:
- am
metrics:
- bertscore
base_model:
- genmo/mochi-1-preview
new_version: Etched/oasis-500m
pipeline_tag: text-generation
---
# Oasis 500M

![](./media/arch.png)

![](./media/thumb.png)

Oasis is an interactive world model developed by [Decart](https://www.decart.ai/) and [Etched](https://www.etched.com/). Based on diffusion transformers, Oasis takes in user keyboard input and generates gameplay in an autoregressive manner. Here we release the weights for Oasis 500M, a downscaled version of the model, along with inference code for action-conditional frame generation.

For more details, see our [joint blog post](https://oasis-model.github.io/) to learn more.

## Run local inference
To generate action-conditional video using the 500M model, please see the instructions on our [inference repo](https://github.com/etched-ai/open-oasis/tree/master)!

And to use the most powerful version of the model, be sure to check out the [live demo](https://oasis.us.decart.ai/)!