https://github.com/alpa-projects/alpa - Release v0.2.3

Maintenance release.

• Support manual sharding (#816)
• Optimize cross-mesh communication (#773, #798)
• Add publications (#885)
• Support new models: CodeGen, BLOOM-Z (#774, #844)
• Add priority scheduling in the model server (#852)
• Add guidance on strategy inspection (#879)
• Inference stage construction (#793, #799)
• Misc bug fixes (#876, #878, #873)

- Python
Published by merrymercy about 3 years ago

https://github.com/alpa-projects/alpa - Release v0.2.2

• Rebase to jax v0.3.22 (#782)
• Support TPU backend for ShardParallel (#764)
• Misc improvements for LLM serving

- Python
Published by merrymercy over 3 years ago

https://github.com/alpa-projects/alpa - Release v0.2.1

• Fine-tune OPT with 3d parallelism #716
• Support bigscience/bloom #707
• Add model serving controller #720
• Jaxlib update

- Python
Published by merrymercy over 3 years ago

https://github.com/alpa-projects/alpa - Release v0.2.0

• Improve OPT generation speed (#661)
• Add logprobs endpoint (#679)
• Support global norm in pipeline parallelism (#672)
• Rebase jax to v0.3.15 (#682)
• Improve the stability of the server

- Python
Published by merrymercy over 3 years ago

https://github.com/alpa-projects/alpa - Release v0.1.6

• Improve OPT-175B demo https://opt.alpa.ai/. Our website has served 20k+ requests during the last week.
• Add OPT fine-tuning examples https://github.com/alpa-projects/alpa/tree/main/examples/opt_finetune
• Various bug fixes

- Python
Published by merrymercy almost 4 years ago

https://github.com/alpa-projects/alpa - Release v0.1.5

• Improve OPT-175B demo https://opt.alpa.ai/
• Improve version management
• Various bug fixes

- Python
Published by merrymercy almost 4 years ago

https://github.com/alpa-projects/alpa - Release v0.1.4

• [Feature] Serialize Parallel Plan (#587)
• [Test] Move test install to alpa package (#584)
• [CI/CD] New PyPI Infra (#577)

- Python
Published by merrymercy almost 4 years ago

https://github.com/alpa-projects/alpa - Release v0.1.3

• [Tutorial] Serve OPT-175B (https://alpa-projects.github.io/tutorials/opt_serving.html)
  • New website
  • Better weight conversion scripts
• [API Change] Move layer construction decorators to ParallelMethod (#565)

- Python
Published by merrymercy almost 4 years ago

https://github.com/alpa-projects/alpa - Release v0.1.1

• Interface change (https://github.com/alpa-projects/alpa/pull/540)
• Minor bug fixes

- Python
Published by merrymercy almost 4 years ago

https://github.com/alpa-projects/alpa - Release v0.1.0

Major feature updates: - Support serving Meta's open source OPT model 175B - Distributed weight init - Add a set of strategy-specific interfaces such as DataParallel, Zero2Parallel, Zero3Parallel

Minors: - enhancement on the PyTorch frontend - many bugfixes - Doc update: Performance tuning guide, and a slide deck for Alpa

- Python
Published by zhisbug almost 4 years ago

https://github.com/alpa-projects/alpa - v0.0.3 release

• PyTorch experimental support by @yf225
• small changes of interfaces on auto sharding

- Python
Published by zhisbug almost 4 years ago

https://github.com/alpa-projects/alpa - Test release v0.0.2

- Python
Published by merrymercy almost 4 years ago

https://github.com/alpa-projects/alpa - First release of v0.0.1

First release of Alpa v0.0.1

- Python
Published by zhisbug almost 4 years ago