[Core] Support load and unload LoRA in api server (#6566)

Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>

vllm/envs.py

@@ -61,6 +61,7 @@ if TYPE_CHECKING:
     VLLM_ALLOW_ENGINE_USE_RAY: bool = False
     VLLM_PLUGINS: Optional[List[str]] = None
     VLLM_TORCH_PROFILER_DIR: Optional[str] = None
+    VLLM_ALLOW_RUNTIME_LORA_UPDATING: bool = False
 
 
 def get_default_cache_root():
@@ -409,6 +410,12 @@ environment_variables: Dict[str, Callable[[], Any]] = {
     # If set, vLLM will use Triton implementations of AWQ.
     "VLLM_USE_TRITON_AWQ":
     lambda: bool(int(os.getenv("VLLM_USE_TRITON_AWQ", "0"))),
+
+    # If set, allow loading or unloading lora adapters in runtime,
+    "VLLM_ALLOW_RUNTIME_LORA_UPDATING":
+    lambda:
+    (os.environ.get("VLLM_ALLOW_RUNTIME_LORA_UPDATING", "0").strip().lower() in
+     ("1", "true")),
 }
 
 # end-env-vars-definition