Skip to content

add more options for jax_default_matmul_precision #6217

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

add more options for jax_default_matmul_precision #6217

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 6 additions & 3 deletions jax/config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -420,7 +420,9 @@ def _update_x64_thread_local(val):

default_matmul_precision = config.define_enum_state(
name='jax_default_matmul_precision',
enum_values=['bfloat16', 'tensorfloat32', 'float32'],
enum_values=['bfloat16', 'fastest',
'tensorfloat32', 'bfloat16_3x',
'float32', 'highest'],
default=None,
help=('Control the default matmul and conv precision for 32bit inputs.\n\n'

Expand All @@ -436,5 +438,6 @@ def _update_x64_thread_local(val):
'level for computations involved in matrix multiplication and '
'convolution on 32bit inputs. The levels roughly describe the '
"precision at which scalar products are computed. The 'bfloat16' "
"option is the fastest and least precise; 'float32' is similar to "
"full float32 precision; 'tensorfloat32' is intermediate.\n\n"))
"option is the fastest and least precise (alias: 'fastest'); "
"'float32' is similar to full float32 precision (alias: 'highest'); "
"'tensorfloat32' is intermediate (alias: 'bfloat16_3x').\n\n"))