Fix empty /v1/models response for locally loaded models (ml-explore#713)

* server: expose local --model in /v1/models.
use full local path as model id for /v1/models

在加载本地模型时，/v1/models返回空列表，导致 open webui获取不到模型，没办法使用。所以这里把本地模型的全路径作为 model_id返回。

* nits

* fix test

---------

Co-authored-by: chenxilong <cxl750529174@163.com>
Co-authored-by: Awni Hannun <awni.hannun@gmail.com>

Author: 3 people

mlx_lm/server.py

@@ -1444,6 +1444,18 @@ def probably_mlx_lm(repo):
             for repo in downloaded_models
         ]
 
+        if self.response_generator.cli_args.model:
+            model_path = Path(self.response_generator.cli_args.model)
+            if model_path.exists():
+                model_id = str(model_path.resolve())
+                models.append(
+                    {
+                        "id": model_id,
+                        "object": "model",
+                        "created": self.created,
+                    }
+                )
+
         response = {"object": "list", "data": models}
 
         response_json = json.dumps(response).encode()

tests/test_server.py

@@ -40,6 +40,7 @@ def __init__(self, with_draft=False):
                 "min_p": 0.0,
                 "max_tokens": 512,
                 "chat_template_args": {},
+                "model": None,
             },
         )