Skip to content

Latest commit

 

History

History
17 lines (11 loc) · 462 Bytes

README.md

File metadata and controls

17 lines (11 loc) · 462 Bytes

lcm_test

Quick and dirty Streamlit UI for Latent Consistency Models.

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.