14s now for 4x4 on A100
This commit is contained in:
parent
8f7f0ba756
commit
672f325b3b
2
README.md
vendored
2
README.md
vendored
|
@ -11,7 +11,7 @@ This is a fast, minimal implementation of Boris Dayma's [DALL·E Mega](https://g
|
||||||
To generate a 4x4 grid of DALL·E Mega images it takes
|
To generate a 4x4 grid of DALL·E Mega images it takes
|
||||||
- 89 sec with a T4 in Colab
|
- 89 sec with a T4 in Colab
|
||||||
- 48 sec with a P100 in Colab
|
- 48 sec with a P100 in Colab
|
||||||
- 16 sec with an A100 on Replicate
|
- 14 sec with an A100 on Replicate
|
||||||
- TBD with an H100 (@NVIDIA?)
|
- TBD with an H100 (@NVIDIA?)
|
||||||
|
|
||||||
The flax model and code for converting it to torch can be found [here](https://github.com/kuprel/min-dalle-flax).
|
The flax model and code for converting it to torch can be found [here](https://github.com/kuprel/min-dalle-flax).
|
||||||
|
|
Loading…
Reference in New Issue
Block a user