haolu7

4 ms·

1. 1.▲
   8bit HippoAttention: Up to 3X Faster Compared to FlashAttentionV2(blog.hippoml.com)

   4 points by haolu7 3y ago|0 comments
2. 2.▲
   by haolu7 4y ago

   No, you don't need to train the model differently to use it with AITemplate. Here is an intro example to do inference with AITemplate with a very simple PyTorch model: https://facebookincubator.github.io/AITemplate&#x2F
3. 3.▲
   by haolu7 4y ago

   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