Documentation: competing frameworks
borgr opened this issue · comments
Stating why one should choose this framework instead of others (GPTneoX DS\megatron, accelerate+HF etc.) may ease the work when choosing to use this framework rather than another. (The timing vs ease of use you started mentioning on Twitter might be a good thing to write there first)
Agree thanks for raising. This is in progress but to provide some updates:
- Added the following sentence to the readme: "In contrast with other repositories such as Megatron, we depend only on PyTorch and Triton (via xformers) for our core modeling code"
- Achal led an effort to get large speed-ups, and the speed is now very close to Mosaic, check out some numbers/details here: #29
Very cool go open LMs go openLM