diff --git a/llama_stack/providers/inline/eval/meta_reference/eval.py b/llama_stack/providers/inline/eval/meta_reference/eval.py
index ba2fc7c95..58241eb42 100644
--- a/llama_stack/providers/inline/eval/meta_reference/eval.py
+++ b/llama_stack/providers/inline/eval/meta_reference/eval.py
@@ -150,7 +150,7 @@ class MetaReferenceEvalImpl(Eval, EvalTasksProtocolPrivate):
                     messages.append(candidate.system_message)
                 messages += input_messages
                 response = await self.inference_api.chat_completion(
-                    model=candidate.model,
+                    model_id=candidate.model,
                     messages=messages,
                     sampling_params=candidate.sampling_params,
                 )
diff --git a/llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/llm_as_judge_scoring_fn.py b/llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/llm_as_judge_scoring_fn.py
index a950f35f9..4b43de93f 100644
--- a/llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/llm_as_judge_scoring_fn.py
+++ b/llama_stack/providers/inline/scoring/llm_as_judge/scoring_fn/llm_as_judge_scoring_fn.py
@@ -62,7 +62,7 @@ class LlmAsJudgeScoringFn(BaseScoringFn):
         )
 
         judge_response = await self.inference_api.chat_completion(
-            model=fn_def.params.judge_model,
+            model_id=fn_def.params.judge_model,
             messages=[
                 {
                     "role": "user",
diff --git a/llama_stack/providers/remote/inference/vllm/vllm.py b/llama_stack/providers/remote/inference/vllm/vllm.py
index 2d03a9ef8..e5eb6e1ea 100644
--- a/llama_stack/providers/remote/inference/vllm/vllm.py
+++ b/llama_stack/providers/remote/inference/vllm/vllm.py
@@ -85,7 +85,6 @@ class VLLMInferenceAdapter(Inference, ModelRegistryHelper, ModelsProtocolPrivate
         logprobs: Optional[LogProbConfig] = None,
     ) -> AsyncGenerator:
         model = await self.model_store.get_model(model_id)
-        print(f"model={model}")
         request = ChatCompletionRequest(
             model=model.provider_resource_id,
             messages=messages,