0.1

Tencent/WeKnora0.1Aug 11, 2023by jafioti

AI Summary

This is the first functional release of Luminal, a deep learning framework dedicated to static computation and aggressive optimization. It introduces the ability to run the LLaMA 1 7B model to generate valid text, albeit at a slow speed.

Key Highlights

  • First functional version of the Luminal framework.
  • Focus on static computation and aggressive optimization.
  • Successfully runs the LLaMA 1 7B model.
  • Generates valid text output.

New Features

  • Static computation deep learning framework.
  • LLaMA 1 7B model inference support.
  • Text generation capabilities.

Full Release Notes

This is the first release of luminal, a deep learning framework focused on static computation and aggressive optimization.

This release marks the first functional version. Currently examples/llama can run the full llama 1 7B and generate valid text, though it runs very slow.

Future work will speed this up significantly.