[Core] Add Support for Default Modality Specific LoRAs [generate / chat completions] (#19126)

Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
2025-07-10 14:09:37 -06:00
parent 3de2ed767f
commit 41060c6e08
9 changed files with 482 additions and 5 deletions
--- a/vllm/entrypoints/openai/serving_chat.py
+++ b/vllm/entrypoints/openai/serving_chat.py
@@ -153,7 +153,8 @@ class OpenAIServingChat(OpenAIServing):
            (
                lora_request,
                prompt_adapter_request,
-            ) = self._maybe_get_adapters(request)
+            ) = self._maybe_get_adapters(request,
+                                         supports_default_mm_loras=True)

            model_name = self._get_model_name(request.model, lora_request)