Skip to content

🚩 Enable MM encoding feature by default for a… #1029

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

Merged
gkumbhat merged 3 commits into from
Jul 14, 2026
+3 −8
Merged

🚩 Enable MM encoding feature by default for a… #1029

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
b0e49c0
:triangular_flag_on_post: Enable MM encoding feature by default for a…
gkumbhat Jul 13, 2026
eddf380
:label: Change default dtype for power to float32
gkumbhat Jul 14, 2026
38f81f6
:white_check_marks: Fix env check
gkumbhat Jul 14, 2026
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: 2 additions & 7 deletions sendnn_inference/envs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@

_cache: dict[str, Any] = {}

_CPU_MM_DTYPE_PLATFORM_DEFAULTS = {"s390x": "float32", "ppc64le": "bfloat16"}
_CPU_MM_DTYPE_PLATFORM_DEFAULTS = {"s390x": "float32", "ppc64le": "float32"}


def override(name: str, value: str) -> None:
Expand Down Expand Up @@ -184,12 +184,7 @@ def clear_env_cache():
# its embedding is available. Only effective for decoder models with TP > 1.
# Defaults to 0 (disabled) — uses the Phase 1 blocking encode path.
"SENDNN_INFERENCE_ASYNC_MM_ENCODER": lambda: bool(
int(
os.getenv(
"SENDNN_INFERENCE_ASYNC_MM_ENCODER",
"0" if platform.machine() == "ppc64le" else "1",
)
)
int(os.getenv("SENDNN_INFERENCE_ASYNC_MM_ENCODER", "1"))
),
# When "1" (default), rank 0 runs the vision encoder and shares the result
# with other TP ranks via POSIX shared memory (one encoder call instead of
Expand Down
2 changes: 1 addition & 1 deletion tests/utils/test_envs.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,7 @@ def test_env_vars_override_for_invalid_config():
"machine,expected",
[
("s390x", torch.float32),
("ppc64le", torch.bfloat16),
("ppc64le", torch.float32),
("x86_64", torch.float16),
("aarch64", torch.float16),
],
Expand Down
Loading