faster inference with cuda/cudnn backends flags

2022-07-09 06:48:51 -04:00
parent 703bfb231d
commit dba3f11b3f
8 changed files with 19 additions and 14 deletions
@@ -11,7 +11,7 @@ This is a fast, minimal port of Boris Dayma's [DALL·E Mega](https://github.com/
 To generate a 4x4 grid of DALL·E Mega images it takes:
 - 89 sec with a T4 in Colab
 - 48 sec with a P100 in Colab
- 14 sec with an A100 on Replicate
+- 13 sec with an A100 on Replicate

 The flax model and code for converting it to torch can be found [here](https://github.com/kuprel/min-dalle-flax).