https://github.com/akx/lcm_test

Quick and dirty Streamlit UI for Latent Consistency Models
https://github.com/akx/lcm_test

generative-art latent-consistency-model stable-diffusion streamlit

Last synced: 3 months ago
JSON representation

Quick and dirty Streamlit UI for Latent Consistency Models

• Host: GitHub
• URL: https://github.com/akx/lcm_test
• Owner: akx
• License: mit
• Created: 2023-10-24T11:16:50.000Z (over 1 year ago)
• Default Branch: master
• Last Pushed: 2023-10-24T12:01:56.000Z (over 1 year ago)
• Last Synced: 2024-10-26T19:58:02.758Z (3 months ago)
• Topics: generative-art, latent-consistency-model, stable-diffusion, streamlit
• Language: Python
• Homepage:
• Size: 5.86 KB
• Stars: 8
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# lcm_test

Quick and dirty Streamlit UI for [Latent Consistency Models](https://github.com/luosiallen/latent-consistency-model).

## Usage

Tested on an M2 Mac, so YMMV, but:

```

pip install -r requirements.txt

streamlit run lcm_test.py

```

## Configuration

* If you want to force a given Torch device, set the `LCM_DEVICE` envvar to e.g. `cpu`.

* If you want to force using fp16 precision, set the `LCM_FP16` envvar to anything truthy. This might not work.