[QeRL] Layerwise Reloading (#32133)

Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
2026-01-30 10:50:05 -05:00
parent 74898a7015
commit f857a03f6b
17 changed files with 923 additions and 314 deletions
--- a/vllm/v1/worker/gpu_worker.py
+++ b/vllm/v1/worker/gpu_worker.py
@@ -280,8 +280,8 @@ class Worker(WorkerBase):
    def update_config(self, overrides: dict[str, Any]) -> None:
        self.model_runner.update_config(overrides)

-    def reload_weights(self) -> None:
-        self.model_runner.reload_weights()
+    def reload_weights(self, *args, **kwargs) -> None:
+        self.model_runner.reload_weights(*args, **kwargs)

    @torch.inference_mode()
    def determine_available_memory(self) -> int: