Hacker GitHub

← All repos

deepspeedai/DeepSpeed

View on GitHub ↗ ★ 43,285 Python

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

18 mentions on Hacker News by 17 people: 10 stories, 8 comments · first 2020-02-10 · latest 2024-11-26

Mentions per year 18 mentions · busiest year 2023 (8)
20082026
AllStoriesComments NewestTopOldest
  1. …[2]: https://arxiv.org/abs/2101.06840 [3]: https://github.com/nvidia/Megatron-LM [4]: https://github.com/microsoft/DeepSpeed/issues/1110
    comment by sighingnow · 2023-10-10 · in Why async gradient update doesn't get popular in LLM community?
  2. …has other advantages.) [1] https://github.com/pytorch/pytorch/issues?q=is%3Aissue+is%3A... [2] https://github.com/microsoft/DeepSpeed/issues/1580 [3] https://github.com/TimDettmers/bitsandbytes/issues/485 [4]…
    comment by lhl · 2023-07-26 · in Guide to running Llama 2 locally
  3. …- https://github.com/bigscience-workshop/petals - https://github.com/FMInference/FlexGen - https://github.com/microsoft/DeepSpeed Anything that could bring this to a 10GB 3080 or 24GB 3090 without 60s/it per token?
    comment by underlines · 2023-03-03 · in Facebook LLAMA is being openly distributed via torrents
  4. …reimplementing everything yourself. Probably FairScale [2] too, but never tried it myself. [1]: https://github.com/microsoft/DeepSpeed [2]: https://github.com/facebookresearch/fairscale
    comment by Voloskaya · 2021-09-27 · in How to train large models on many GPUs?
  5. …contributing to the ML ecosystem already by the sole virtue that it isn't a Google product. [1] = https://github.com/microsoft/DeepSpeed [2] = https://pytorch.org/docs/stable/jit.html
    comment by dwrodri · 2021-04-16 · in Flashlight: Fast and flexible machine learning in C++
  6. …wrong. I haven’t had time to implement the fix but I don’t see any reason to assume it won’t work. https://github.com/microsoft/DeepSpeed/issues/846 Also, the specific problem described in that Issue was due to a bug I found in DeepSpeed…
    comment by stellaathena · 2021-03-13 · in Zero-3 Offload: Scale DL models to trillion parameters without code changes
  7. The code for the distributed training library, not the model - https://github.com/microsoft/DeepSpeed/
    comment by Tenoke · 2020-02-10 · in Turing-NLG: A 17B-parameter language model
  8. Link to GitHub: https://github.com/microsoft/DeepSpeed
    comment by minimaxir · 2020-02-10 · in Microsoft Zero and DeepSpeed: Memory Efficient Large Neural Network Training