add test cases

Former-commit-id: 731176ff34cdf0cbf6b41c40c69f4ceb54c2daf6

src/llamafactory/hparams/model_args.py

@@ -136,10 +136,6 @@ class ModelArguments:
         default=8,
         metadata={"help": "Maximum rank of all LoRAs in the vLLM engine."},
     )
-    vllm_dtype: Literal["auto", "float16", "bfloat16", "float32"] = field(
-        default="auto",
-        metadata={"help": "Data type for model weights and activations in the vLLM engine."},
-    )
     offload_folder: str = field(
         default="offload",
         metadata={"help": "Path to offload model weights."},
@@ -148,6 +144,10 @@ class ModelArguments:
         default=True,
         metadata={"help": "Whether or not to use KV cache in generation."},
     )
+    infer_dtype: Literal["auto", "float16", "bfloat16", "float32"] = field(
+        default="auto",
+        metadata={"help": "Data type for model weights and activations at inference."}
+    )
     hf_hub_token: Optional[str] = field(
         default=None,
         metadata={"help": "Auth token to log in with Hugging Face Hub."},