Skip to content

Enable Pipeline Parallelism to use mp as distributed backend on Jax TPU platform #1054

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
merged 1 commit into from
Nov 14, 2025
Merged

Enable Pipeline Parallelism to use mp as distributed backend on Jax TPU platform #1054

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
10 changes: 8 additions & 2 deletions tpu_inference/platforms/tpu_platform.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -184,8 +184,14 @@ def check_and_update_config(cls, vllm_config: VllmConfig) -> None:

multihost_backend = os.environ.get("TPU_MULTIHOST_BACKEND", "").lower()
if not multihost_backend: # Single host
logger.info("Force using UniProcExecutor for JAX on single host.")
parallel_config.distributed_executor_backend = "uni"
if parallel_config.pipeline_parallel_size == 1:
logger.info("Force using UniProcExecutor for JAX on \
single host without pipeline parallelism.")
parallel_config.distributed_executor_backend = "uni"
else:
logger.info("Force using MultiprocExecutor for JAX on \
single host with pipeline parallelism.")
parallel_config.distributed_executor_backend = "mp"
elif multihost_backend == "ray":
from tpu_inference.executors.ray_distributed_executor import \
RayDistributedExecutor
Expand Down