Skip to content

[Raiden Weight Sync 4/7] Target-free and streaming weight conversion for vLLM #5169

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
copybara-service merged 1 commit into from
Sep 10, 2026
+609 −67
Merged

[Raiden Weight Sync 4/7] Target-free and streaming weight conversion for vLLM #5169

Show file tree
Hide file tree
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
16 changes: 16 additions & 0 deletions src/maxtext/integration/vllm/convert_utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -689,3 +689,19 @@ def _scanned_sharding_from_per_layer(
jax.sharding.PartitionSpec(*spec),
memory_kind=per_layer_sharding.memory_kind,
)


def resolve_rollout_tp(config: Any, tp: int = 1) -> int:
"""Resolves rollout TP from config."""
if tp > 1:
return int(tp)

config_tp = 0
if config is not None:
config_tp = int(
getattr(config, "rollout_tensor_parallelism", 0)
or getattr(getattr(config, "cluster", None), "rollout_tensor_parallelism", 0)
or getattr(config, "rollout_mesh_tp", 0)
or 0
)
return int(config_tp or 1)
Loading
Loading