Document Sync by Tina

Author: Chivier

docs/stable/cli/cli_api.md

@@ -62,16 +62,16 @@ sllm-cli deploy [OPTIONS]
 - `--backend <backend_name>`
   - Overwrite the backend in the default configuration.
 
-- `--num_gpus <number>`
+- `--num-gpus <number>`
   - Overwrite the number of GPUs in the default configuration.
 
 - `--target <number>`
   - Overwrite the target concurrency in the default configuration.
 
-- `--min_instances <number>`
+- `--min-instances <number>`
   - Overwrite the minimum instances in the default configuration.
 
-- `--max_instances <number>`
+- `--max-instances <number>`
   - Overwrite the maximum instances in the default configuration.
 
 ##### Examples
@@ -92,7 +92,7 @@ sllm-cli deploy --model facebook/opt-1.3b --backend transformers
 
 Deploy using a model name and overwrite multiple configurations:
 ```bash
-sllm-cli deploy --model facebook/opt-1.3b --num_gpus 2 --target 5 --min_instances 1 --max_instances 5
+sllm-cli deploy --model facebook/opt-1.3b --num-gpus 2 --target 5 --min-instances 1 --max-instances 5
 ```
 
 ##### Example Configuration File (`config.json`)
@@ -275,15 +275,15 @@ sllm-cli fine-tuning [OPTIONS]
 ```
 
 ##### Options
-- `--base_model <model_name>`
+- `--base-model <model_name>`
   - Base model name to be fine-tuned
 - `--config <config_path>`
   - Path to the JSON configuration file.
 
 ##### Example
 ```bash
-sllm-cli fine-tuning --base_model <model_name>
-sllm-cli fine-tuning --base_model <model_name> --config <path_to_ft_config_file>
+sllm-cli fine-tuning --base-model <model_name>
+sllm-cli fine-tuning --base-model <model_name> --config <path_to_ft_config_file>
 ```
 
 ##### Example Configuration File (`ft_config.json`)

docs/stable/serve/storage_aware_scheduling.md

@@ -33,7 +33,7 @@ Replace `/path/to/your/models` with the actual path where you want to store the
 
 ### Step 3: Enable Storage Aware Scheduling in Docker Compose
 
-The Docker Compose configuration is already located in the `examples/storage_aware_scheduling` directory. To activate storage-aware scheduling, ensure the `docker-compose.yml` file includes the necessary configurations(`sllm_head` service should include the `--enable_storage_aware` command).
+The Docker Compose configuration is already located in the `examples/storage_aware_scheduling` directory. To activate storage-aware scheduling, ensure the `docker-compose.yml` file includes the necessary configurations(`sllm_head` service should include the `--enable-storage-aware` command).
 
 :::tip
 Recommend to adjust the number of GPUs and `mem_pool_size` based on the resources available on your machine.

docs/stable/store/quickstart.md

@@ -134,11 +134,11 @@ Thus, for fist-time users, you have to load the model from other backends and th
 
 1. Download the model from HuggingFace and save it in the ServerlessLLM format:
 ``` bash
-python3 examples/sllm_store/save_vllm_model.py --model_name facebook/opt-1.3b --storage_path $PWD/models --tensor_parallel_size 1
+python3 examples/sllm_store/save_vllm_model.py --model-name facebook/opt-1.3b --storage-path $PWD/models --tensor-parallel-size 1
 
 ```
 
-You can also transfer the model from the local path compared to download it from network by passing the `--local_model_path` argument.
+You can also transfer the model from the local path compared to download it from network by passing the `--local-model-path` argument.
 
 After downloading the model, you can launch the checkpoint store server and load the model in vLLM through `sllm` load format.
 

docs/stable/store/rocm_quickstart.md

@@ -65,8 +65,8 @@ docker exec -it sllm_store_server /bin/bash
 Try to save and load a transformer model:
 
 ``` bash
-python3 examples/save_transformers_model.py --model_name "facebook/opt-1.3b" --storage_path "/models"
-python3 examples/load_transformers_model.py --model_name "facebook/opt-1.3b" --storage_path "/models"
+python3 examples/save_transformers_model.py --model-name "facebook/opt-1.3b" --storage-path "/models"
+python3 examples/load_transformers_model.py --model-name "facebook/opt-1.3b" --storage-path "/models"
 ```
 Expected output: