Create executables for generate, lora, server, merge, convert (#682)

* feat: create executables mlx_lm.<cmd>

* nits in docs

---------

Co-authored-by: Awni Hannun <awni@apple.com>

llms/README.md

@@ -66,7 +66,7 @@ To see a description of all the arguments you can do:
 You can also use `mlx-lm` from the command line with:
 
 ```
-python -m mlx_lm.generate --model mistralai/Mistral-7B-Instruct-v0.1 --prompt "hello"
+mlx_lm.generate --model mistralai/Mistral-7B-Instruct-v0.1 --prompt "hello"
 ```
 
 This will download a Mistral 7B model from the Hugging Face Hub and generate
@@ -75,19 +75,19 @@ text using the given prompt.
 For a full list of options run:
 
 ```
-python -m mlx_lm.generate --help
+mlx_lm.generate --help
 ```
 
 To quantize a model from the command line run:
 
 ```
-python -m mlx_lm.convert --hf-path mistralai/Mistral-7B-Instruct-v0.1 -q
+mlx_lm.convert --hf-path mistralai/Mistral-7B-Instruct-v0.1 -q
 ```
 
 For more options run:
 
 ```
-python -m mlx_lm.convert --help
+mlx_lm.convert --help
 ```
 
 You can upload new models to Hugging Face by specifying `--upload-repo` to
@@ -95,7 +95,7 @@ You can upload new models to Hugging Face by specifying `--upload-repo` to
 [MLX Hugging Face community](https://huggingface.co/mlx-community) you can do:
 
 ```
-python -m mlx_lm.convert \
+mlx_lm.convert \
     --hf-path mistralai/Mistral-7B-v0.1 \
     -q \
     --upload-repo mlx-community/my-4bit-mistral

llms/mlx_lm/LORA.md

@@ -27,7 +27,7 @@ LoRA (QLoRA).[^qlora] LoRA fine-tuning works with the following model families:
 The main command is `mlx_lm.lora`. To see a full list of command-line options run:
 
 ```shell
-python -m mlx_lm.lora --help
+mlx_lm.lora --help
 ```
 
 Note, in the following the `--model` argument can be any compatible Hugging
@@ -37,7 +37,7 @@ You can also specify a YAML config with `-c`/`--config`. For more on the format
 [example YAML](examples/lora_config.yaml). For example:
 
 ```shell
-python -m mlx_lm.lora --config /path/to/config.yaml
+mlx_lm.lora --config /path/to/config.yaml
 ```
 
 If command-line flags are also used, they will override the corresponding
@@ -48,7 +48,7 @@ values in the config.
 To fine-tune a model use:
 
 ```shell
-python -m mlx_lm.lora \
+mlx_lm.lora \
     --model <path_to_model> \
     --train \
     --data <path_to_data> \
@@ -76,7 +76,7 @@ You can resume fine-tuning with an existing adapter with
 To compute test set perplexity use:
 
 ```shell
-python -m mlx_lm.lora \
+mlx_lm.lora \
     --model <path_to_model> \
     --adapter-path <path_to_adapters> \
     --data <path_to_data> \
@@ -88,7 +88,7 @@ python -m mlx_lm.lora \
 For generation use `mlx_lm.generate`:
 
 ```shell
-python -m mlx_lm.generate \
+mlx_lm.generate \
     --model <path_to_model> \
     --adapter-path <path_to_adapters> \
     --prompt "<your_model_prompt>"
@@ -106,13 +106,13 @@ You can generate a model fused with the low-rank adapters using the
 To see supported options run:
 
 ```shell
-python -m mlx_lm.fuse --help
+mlx_lm.fuse --help
 ```
 
 To generate the fused model run:
 
 ```shell
-python -m mlx_lm.fuse --model <path_to_model>
+mlx_lm.fuse --model <path_to_model>
 ```
 
 This will by default load the adapters from `adapters/`, and save the fused
@@ -125,7 +125,7 @@ useful for the sake of attribution and model versioning.
 For example, to fuse and upload a model derived from Mistral-7B-v0.1, run: 
 
 ```shell
-python -m mlx_lm.fuse \
+mlx_lm.fuse \
     --model mistralai/Mistral-7B-v0.1 \
     --upload-repo mlx-community/my-4bit-lora-mistral \
     --hf-path mistralai/Mistral-7B-v0.1
@@ -134,7 +134,7 @@ python -m mlx_lm.fuse \
 To export a fused model to GGUF, run:
 
 ```shell
-python -m mlx_lm.fuse \
+mlx_lm.fuse \
     --model mistralai/Mistral-7B-v0.1 \
     --export-gguf
 ```

llms/mlx_lm/MERGE.md

@@ -6,14 +6,14 @@ Face hub or save them locally for LoRA fine tuning.
 The main command is `mlx_lm.merge`:
 
 ```shell
-python -m mlx_lm.merge --config config.yaml 
+mlx_lm.merge --config config.yaml 
 ```
 
 The merged model will be saved by default in `mlx_merged_model`. To see a
 full list of options run:
 
 ```shell
-python -m mlx_lm.merge --help
+mlx_lm.merge --help
 ```
 
 Here is an example `config.yaml`:

llms/mlx_lm/SERVER.md

@@ -11,13 +11,13 @@ API](https://platform.openai.com/docs/api-reference).
 Start the server with: 
 
 ```shell
-python -m mlx_lm.server --model <path_to_model_or_hf_repo>
+mlx_lm.server --model <path_to_model_or_hf_repo>
 ```
 
 For example:
 
 ```shell
-python -m mlx_lm.server --model mistralai/Mistral-7B-Instruct-v0.1
+mlx_lm.server --model mistralai/Mistral-7B-Instruct-v0.1
 ```
 
 This will start a text generation server on port `8080` of the `localhost`
@@ -27,7 +27,7 @@ Hugging Face repo if it is not already in the local cache.
 To see a full list of options run:
 
 ```shell
-python -m mlx_lm.server --help
+mlx_lm.server --help
 ```
 
 You can make a request to the model by running:

llms/mlx_lm/convert.py

@@ -52,7 +52,11 @@ def configure_parser() -> argparse.ArgumentParser:
     return parser
 
 
-if __name__ == "__main__":
+def main():
     parser = configure_parser()
     args = parser.parse_args()
     convert(**vars(args))
+
+
+if __name__ == "__main__":
+    main()

llms/mlx_lm/generate.py

@@ -101,7 +101,10 @@ def colorprint_by_t0(s, t0):
     colorprint(color, s)
 
 
-def main(args):
+def main():
+    parser = setup_arg_parser()
+    args = parser.parse_args()
+
     mx.random.seed(args.seed)
 
     # Building tokenizer_config
@@ -143,6 +146,4 @@ def main(args):
 
 
 if __name__ == "__main__":
-    parser = setup_arg_parser()
+    main()
-    args = parser.parse_args()
-    main(args)

llms/mlx_lm/lora.py

@@ -247,7 +247,7 @@ def run(args, training_callback: TrainingCallback = None):
         print(f"Test loss {test_loss:.3f}, Test ppl {test_ppl:.3f}.")
 
 
-if __name__ == "__main__":
+def main():
     parser = build_parser()
     args = parser.parse_args()
     config = args.config
@@ -266,3 +266,7 @@ if __name__ == "__main__":
         if not args.get(k, None):
             args[k] = v
     run(types.SimpleNamespace(**args))
+
+
+if __name__ == "__main__":
+    main()

llms/mlx_lm/merge.py

@@ -162,7 +162,11 @@ def merge(
         upload_to_hub(mlx_path, upload_repo, base_hf_path)
 
 
-if __name__ == "__main__":
+def main():
     parser = configure_parser()
     args = parser.parse_args()
     merge(**vars(args))
+
+
+if __name__ == "__main__":
+    main()

llms/mlx_lm/server.py

@@ -409,7 +409,7 @@ def run(host: str, port: int, server_class=HTTPServer, handler_class=APIHandler)
     httpd.serve_forever()
 
 
-if __name__ == "__main__":
+def main():
     parser = argparse.ArgumentParser(description="MLX Http Server.")
     parser.add_argument(
         "--model",
@@ -449,3 +449,7 @@ if __name__ == "__main__":
     )
 
     run(args.host, args.port)
+
+
+if __name__ == "__main__":
+    main()

llms/setup.py

@@ -24,4 +24,14 @@ setup(
     install_requires=requirements,
     packages=["mlx_lm", "mlx_lm.models", "mlx_lm.tuner"],
     python_requires=">=3.8",
+    entry_points={
+        "console_scripts": [
+            "mlx_lm.convert = mlx_lm.convert:main",
+            "mlx_lm.fuse = mlx_lm.fuse:main",
+            "mlx_lm.generate = mlx_lm.generate:main",
+            "mlx_lm.lora = mlx_lm.lora:main",
+            "mlx_lm.merge = mlx_lm.merge:main",
+            "mlx_lm.server = mlx_lm.server:main",
+        ]
+    },
 )