fix: simplified the command to use the config directly

Author: TimPietruskyRunPod

scripts/start_vllm.sh

@@ -2,103 +2,28 @@
 set -e
 
 # vLLM Server Startup Script
-# Usage: ./start_vllm.sh [config.yaml|model_path] [additional_args...]
+# Usage: ./start_vllm.sh config.yaml [additional_args...]
 
-show_usage() {
-    echo "Usage: $0 [config.yaml|model_path] [additional_args...]"
+if [ $# -eq 0 ]; then
+    echo "Usage: $0 config.yaml [additional_args...]"
     echo ""
-    echo "Examples:"
-    echo "  # Using YAML config file:"
+    echo "Example:"
     echo "  $0 vllm_config.yaml"
-    echo ""
-    echo "  # Using model path with args:"
-    echo "  $0 ./outputs/lora-out --lora-modules lora_name=./outputs/lora-out"
-    echo "  $0 NousResearch/Llama-3.2-1B --max-model-len 4096"
-    echo "  $0 ./outputs/merged-model --port 8000"
-    echo ""
-    echo "YAML config example:"
-    echo "  model: ./outputs/my-model"
-    echo "  max_model_len: 32768"
-    echo "  gpu_memory_utilization: 0.95"
-    echo "  port: 8000"
-    echo "  host: 0.0.0.0"
-}
-
-if [ $# -eq 0 ]; then
-    show_usage
     exit 1
 fi
 
-INPUT="$1"
-shift  # Remove first arg, keep the rest
+CONFIG_FILE="$1"
+shift  # Remove config file from args, keep additional args
 
-# Check if input is a YAML file
-if [[ "$INPUT" == *.yaml ]] || [[ "$INPUT" == *.yml ]]; then
-    if [ ! -f "$INPUT" ]; then
-        echo "❌ Config file not found: $INPUT"
-        exit 1
-    fi
-    
-    echo "🚀 Starting vLLM server with config file..."
-    echo "📄 Config: $INPUT"
-    
-    # Extract key info from YAML for display
-    if command -v python3 >/dev/null 2>&1; then
-        MODEL=$(python3 -c "
-import yaml
-try:
-    with open('$INPUT', 'r') as f:
-        config = yaml.safe_load(f)
-    print(config.get('model', 'Not specified'))
-except:
-    print('Could not parse config')
-" 2>/dev/null || echo "Could not parse config")
-        
-        PORT=$(python3 -c "
-import yaml
-try:
-    with open('$INPUT', 'r') as f:
-        config = yaml.safe_load(f)
-    print(config.get('port', 8000))
-except:
-    print('8000')
-" 2>/dev/null || echo "8000")
-        
-        HOST=$(python3 -c "
-import yaml
-try:
-    with open('$INPUT', 'r') as f:
-        config = yaml.safe_load(f)
-    print(config.get('host', '0.0.0.0'))
-except:
-    print('0.0.0.0')
-" 2>/dev/null || echo "0.0.0.0")
-        
-        echo "📁 Model: $MODEL"
-        echo "🌐 Server will be available at: http://$HOST:$PORT"
-    fi
-    echo "🔧 Additional args: $*"
-    echo ""
-    
-    # Start vLLM with config file (using dedicated venv)
-    /opt/vllm-venv/bin/python -m vllm.entrypoints.openai.api_server \
-        --config "$INPUT" \
-        "$@"
-        
-else
-    # Treat as model path (legacy mode)
-    MODEL_PATH="$INPUT"
-    
-    echo "🚀 Starting vLLM server..."
-    echo "📁 Model: $MODEL_PATH"
-    echo "🔧 Additional args: $*"
-    echo "🌐 Server will be available at: http://0.0.0.0:8000"
-    echo ""
-    
-    # Start vLLM with the provided model and any additional arguments (using dedicated venv)
-    /opt/vllm-venv/bin/python -m vllm.entrypoints.openai.api_server \
-        --model "$MODEL_PATH" \
-        --host 0.0.0.0 \
-        --port 8000 \
-        "$@"
+if [ ! -f "$CONFIG_FILE" ]; then
+    echo "❌ Config file not found: $CONFIG_FILE"
+    exit 1
 fi
+
+echo "🚀 Starting vLLM server..."
+echo "📄 Config: $CONFIG_FILE"
+echo "🔧 Additional args: $*"
+echo ""
+
+# Start vLLM with config file (using dedicated venv)
+/opt/vllm-venv/bin/vllm serve --config "$CONFIG_FILE" "$@"