[HN Gopher] AITemplate, a revolutionary new inference engine by ...
___________________________________________________________________
AITemplate, a revolutionary new inference engine by Meta AI
Author : azurezyq
Score : 41 points
Date : 2022-10-03 16:10 UTC (6 hours ago)
(HTM) web link (ai.facebook.com)
(TXT) w3m dump (ai.facebook.com)
| ipiszy wrote:
| tl;dr:
|
| Meta is open sourcing AITemplate, an inference engine for both
| Nvidia and AMD GPUs. Code:
| https://github.com/facebookincubator/AITemplate.
|
| AITemplate delivers much better perf (1.9x ~ 12.8x) compared to
| PyTorch eager on SOTA models, including Bert, ResNet, VIT and
| StableDiffusion.
|
| AITemplate also delivers high perf numbers using AMD GPUs
| (MI-250). With AITemplate, MI-250 achieves 80% ~ 96% A100 perf on
| various ResNet / Bert / VIT models.
|
| AITemplate uses sophisticated fusion techniques to optimize perf,
| including vertical, horizontal, and memory fusions.
|
| btw, I'm one of the authors of AITemplate, happy to answer any
| questions.
| throwaway81523 wrote:
| Thanks, that is very helpful. Do you have to train the model
| differently for use with AITemplate? Could it be helpful for
| Leela Chess Zero (LC0)? I think LC0 has a generic Pytorch
| backend, that is several times slower than its NVidia specific
| CUDA backend. I'm not very clueful about this stuff though.
| devcat wrote:
| Sadly it doesn't have Apple GPU backend
| throwaway81523 wrote:
| Tldr?
| house_road wrote:
| It supports both Nvidia and AMD, and both got pretty good
| speedup. This is a great achievement!
| ghoomketu wrote:
| For all the hate that Facebook gets their only redeeming quality
| is these open source projects they have been releasing all along.
|
| Maybe this is to attract better engineers but all in all this has
| been a net postive for software development. So credit where it
| is due.
| azurezyq wrote:
| https://github.com/facebookincubator/AITemplate
| yinghai83 wrote:
| Very impressive results!
| haolu7 wrote:
| AITemplate-PyTorch Stable Diffusion is the fastest stable
| diffusion inference solution by pushing image generation below
| one second on A100 (batch 1: 0.7s / 25 steps, 1.3s / 50 steps;
| batch 3: 1.6s / 25 steps, per image 0.55s; batch 16 7.9s / 25
| steps, per image 0.49s) for the first time, 2.57X faster than
| Keras' XLA-based GPU compilation solution.
|
| More benchmark numbers and repro at:
| https://github.com/facebookincubator/AITemplate/tree/main/ex...
| enoch2090 wrote:
| How would this perform compared with Tensorflow?
___________________________________________________________________
(page generated 2022-10-03 23:02 UTC)