Support GPU pinning for LoRA (#8697)

2025-08-06 19:39:45 -07:00
parent 6ad6c8c9e6
commit 6210e2c4f0
13 changed files with 425 additions and 134 deletions
@@ -311,3 +311,6 @@ class TpModelWorker:
    def unload_lora_adapter(self, recv_req: UnloadLoRAAdapterReqInput):
        result = self.model_runner.unload_lora_adapter(recv_req.to_ref())
        return result
+
+    def can_run_lora_batch(self, lora_ids: list[str]) -> bool:
+        return self.model_runner.lora_manager.validate_lora_batch(lora_ids)