diff --git a/api/src/builds/models.py b/api/src/builds/models.py
index 929dc47..bfd52ad 100644
--- a/api/src/builds/models.py
+++ b/api/src/builds/models.py
@@ -337,11 +337,13 @@ def recursive_munch(d):
     else:
         return d
 
-def build_model(path, device):
+async def build_model(path, device):
+    from ..core.paths import load_json, load_model_weights
+    
     config = Path(__file__).parent / 'config.json'
     assert config.exists(), f'Config path incorrect: config.json not found at {config}'
-    with open(config, 'r') as r:
-        args = recursive_munch(json.load(r))
+    
+    args = recursive_munch(await load_json(config))
     assert args.decoder.type == 'istftnet', f'Unknown decoder type: {args.decoder.type}'
     decoder = Decoder(dim_in=args.hidden_dim, style_dim=args.style_dim, dim_out=args.n_mels,
             resblock_kernel_sizes = args.decoder.resblock_kernel_sizes,
@@ -365,7 +367,8 @@ def build_model(path, device):
         decoder=decoder.to(device).eval(),
         text_encoder=text_encoder.to(device).eval(),
     )
-    for key, state_dict in torch.load(path, map_location='cpu', weights_only=True)['net'].items():
+    weights = await load_model_weights(path, device='cpu')
+    for key, state_dict in weights['net'].items():
         assert key in model, key
         try:
             model[key].load_state_dict(state_dict)
diff --git a/api/src/core/config.py b/api/src/core/config.py
index 3f44350..91aec63 100644
--- a/api/src/core/config.py
+++ b/api/src/core/config.py
@@ -13,10 +13,15 @@ class Settings(BaseSettings):
     output_dir: str = "output"
     output_dir_size_limit_mb: float = 500.0  # Maximum size of output directory in MB
     default_voice: str = "af"
-    model_dir: str = "/app/models"  # Base directory for model files
-    pytorch_model_path: str = "kokoro-v0_19.pth"
-    onnx_model_path: str = "kokoro-v0_19.onnx"
-    voices_dir: str = "voices"
+    use_gpu: bool = False  # Whether to use GPU acceleration if available
+    use_onnx: bool = True  # Whether to use ONNX runtime
+    # Paths relative to api directory
+    model_dir: str = "src/models"  # Model directory relative to api/
+    voices_dir: str = "src/voices"  # Voices directory relative to api/
+    
+    # Model filenames
+    pytorch_model_file: str = "kokoro-v0_19.pth"
+    onnx_model_file: str = "kokoro-v0_19.onnx"
     sample_rate: int = 24000
     max_chunk_size: int = 300  # Maximum size of text chunks for processing
     gap_trim_ms: int = 250  # Amount to trim from streaming chunk ends in milliseconds
@@ -28,6 +33,12 @@ class Settings(BaseSettings):
     onnx_optimization_level: str = "all"  # all, basic, or disabled
     onnx_memory_pattern: bool = True  # Enable memory pattern optimization
     onnx_arena_extend_strategy: str = "kNextPowerOfTwo"  # Memory allocation strategy
+    
+    # ONNX GPU Settings
+    onnx_device_id: int = 0  # GPU device ID to use
+    onnx_gpu_mem_limit: float = 0.7  # Limit GPU memory usage to 70%
+    onnx_cudnn_conv_algo_search: str = "EXHAUSTIVE"  # CUDNN convolution algorithm search
+    onnx_do_copy_in_default_stream: bool = True  # Copy in default CUDA stream
 
     class Config:
         env_file = ".env"
diff --git a/api/src/core/model_config.py b/api/src/core/model_config.py
new file mode 100644
index 0000000..dc82aee
--- /dev/null
+++ b/api/src/core/model_config.py
@@ -0,0 +1,109 @@
+"""Model configuration schemas."""
+
+from pydantic import BaseModel, Field
+
+
+class ONNXCPUConfig(BaseModel):
+    """ONNX CPU runtime configuration."""
+    
+    num_threads: int = Field(8, description="Number of threads for parallel operations")
+    inter_op_threads: int = Field(4, description="Number of threads for operator parallelism")
+    execution_mode: str = Field("parallel", description="ONNX execution mode")
+    optimization_level: str = Field("all", description="ONNX optimization level")
+    memory_pattern: bool = Field(True, description="Enable memory pattern optimization")
+    arena_extend_strategy: str = Field("kNextPowerOfTwo", description="Memory arena strategy")
+
+    class Config:
+        frozen = True
+
+
+class ONNXGPUConfig(ONNXCPUConfig):
+    """ONNX GPU-specific configuration."""
+    
+    device_id: int = Field(0, description="CUDA device ID")
+    gpu_mem_limit: float = Field(0.7, description="Fraction of GPU memory to use")
+    cudnn_conv_algo_search: str = Field("EXHAUSTIVE", description="CuDNN convolution algorithm search")
+    do_copy_in_default_stream: bool = Field(True, description="Copy in default CUDA stream")
+
+    class Config:
+        frozen = True
+
+
+class PyTorchCPUConfig(BaseModel):
+    """PyTorch CPU backend configuration."""
+    
+    max_batch_size: int = Field(32, description="Maximum batch size for batched inference")
+    stream_buffer_size: int = Field(8, description="Size of stream buffer")
+    memory_threshold: float = Field(0.8, description="Memory threshold for cleanup")
+    retry_on_oom: bool = Field(True, description="Whether to retry on OOM errors")
+    num_threads: int = Field(8, description="Number of threads for parallel operations")
+    pin_memory: bool = Field(True, description="Whether to pin memory for faster CPU-GPU transfer")
+
+    class Config:
+        frozen = True
+
+
+class PyTorchGPUConfig(BaseModel):
+    """PyTorch GPU backend configuration."""
+    
+    device_id: int = Field(0, description="CUDA device ID")
+    use_fp16: bool = Field(True, description="Whether to use FP16 precision")
+    use_triton: bool = Field(True, description="Whether to use Triton for CUDA kernels")
+    max_batch_size: int = Field(32, description="Maximum batch size for batched inference")
+    stream_buffer_size: int = Field(8, description="Size of CUDA stream buffer")
+    memory_threshold: float = Field(0.8, description="Memory threshold for cleanup")
+    retry_on_oom: bool = Field(True, description="Whether to retry on OOM errors")
+    sync_cuda: bool = Field(True, description="Whether to synchronize CUDA operations")
+
+    class Config:
+        frozen = True
+    """PyTorch CPU-specific configuration."""
+    
+    num_threads: int = Field(8, description="Number of threads for parallel operations")
+    pin_memory: bool = Field(True, description="Whether to pin memory for faster CPU-GPU transfer")
+
+    class Config:
+        frozen = True
+
+
+class ModelConfig(BaseModel):
+    """Model configuration."""
+    
+    # General settings
+    model_type: str = Field("pytorch", description="Model type ('pytorch' or 'onnx')")
+    device_type: str = Field("auto", description="Device type ('cpu', 'gpu', or 'auto')")
+    cache_models: bool = Field(True, description="Whether to cache loaded models")
+    cache_voices: bool = Field(True, description="Whether to cache voice tensors")
+    voice_cache_size: int = Field(10, description="Maximum number of cached voices")
+    
+    # Backend-specific configs
+    onnx_cpu: ONNXCPUConfig = Field(default_factory=ONNXCPUConfig)
+    onnx_gpu: ONNXGPUConfig = Field(default_factory=ONNXGPUConfig)
+    pytorch_cpu: PyTorchCPUConfig = Field(default_factory=PyTorchCPUConfig)
+    pytorch_gpu: PyTorchGPUConfig = Field(default_factory=PyTorchGPUConfig)
+
+    class Config:
+        frozen = True
+
+    def get_backend_config(self, backend_type: str):
+        """Get configuration for specific backend.
+        
+        Args:
+            backend_type: Backend type ('pytorch_cpu', 'pytorch_gpu', 'onnx_cpu', 'onnx_gpu')
+            
+        Returns:
+            Backend-specific configuration
+            
+        Raises:
+            ValueError: If backend type is invalid
+        """
+        if backend_type not in {
+            'pytorch_cpu', 'pytorch_gpu', 'onnx_cpu', 'onnx_gpu'
+        }:
+            raise ValueError(f"Invalid backend type: {backend_type}")
+            
+        return getattr(self, backend_type)
+
+
+# Global instance
+model_config = ModelConfig()
\ No newline at end of file
diff --git a/api/src/core/paths.py b/api/src/core/paths.py
index d9096a7..3124ece 100644
--- a/api/src/core/paths.py
+++ b/api/src/core/paths.py
@@ -1,9 +1,10 @@
 """Async file and path operations."""
 
 import io
+import json
 import os
 from pathlib import Path
-from typing import List, Optional, AsyncIterator, Callable, Set
+from typing import List, Optional, AsyncIterator, Callable, Set, Dict, Any
 
 import aiofiles
 import aiofiles.os
@@ -87,10 +88,18 @@ async def get_model_path(model_name: str) -> str:
     Raises:
         RuntimeError: If model not found
     """
-    search_paths = [
-        settings.model_dir,
-        os.path.join(os.path.dirname(__file__), "..", "..", "..", "models")
-    ]
+    # Get api directory path (two levels up from core)
+    api_dir = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
+    
+    # Construct model directory path relative to api directory
+    model_dir = os.path.join(api_dir, settings.model_dir)
+    
+    # Ensure model directory exists
+    os.makedirs(model_dir, exist_ok=True)
+    
+    # Search in model directory
+    search_paths = [model_dir]
+    logger.debug(f"Searching for model in path: {model_dir}")
     
     return await _find_file(model_name, search_paths)
 
@@ -107,12 +116,20 @@ async def get_voice_path(voice_name: str) -> str:
     Raises:
         RuntimeError: If voice not found
     """
+    # Get api directory path
+    api_dir = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
+    
+    # Construct voice directory path relative to api directory
+    voice_dir = os.path.join(api_dir, settings.voices_dir)
+    
+    # Ensure voice directory exists
+    os.makedirs(voice_dir, exist_ok=True)
+    
     voice_file = f"{voice_name}.pt"
     
-    search_paths = [
-        os.path.join(settings.model_dir, "..", settings.voices_dir),
-        os.path.join(os.path.dirname(__file__), "..", settings.voices_dir)
-    ]
+    # Search in voice directory
+    search_paths = [voice_dir]
+    logger.debug(f"Searching for voice in path: {voice_dir}")
     
     return await _find_file(voice_file, search_paths)
 
@@ -123,10 +140,18 @@ async def list_voices() -> List[str]:
     Returns:
         List of voice names (without .pt extension)
     """
-    search_paths = [
-        os.path.join(settings.model_dir, "..", settings.voices_dir),
-        os.path.join(os.path.dirname(__file__), "..", settings.voices_dir)
-    ]
+    # Get api directory path
+    api_dir = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
+    
+    # Construct voice directory path relative to api directory
+    voice_dir = os.path.join(api_dir, settings.voices_dir)
+    
+    # Ensure voice directory exists
+    os.makedirs(voice_dir, exist_ok=True)
+    
+    # Search in voice directory
+    search_paths = [voice_dir]
+    logger.debug(f"Scanning for voices in path: {voice_dir}")
     
     def filter_voice_files(name: str) -> bool:
         return name.endswith('.pt')
@@ -179,6 +204,51 @@ async def save_voice_tensor(tensor: torch.Tensor, voice_path: str) -> None:
         raise RuntimeError(f"Failed to save voice tensor to {voice_path}: {e}")
 
 
+async def load_json(path: str) -> dict:
+    """Load JSON file asynchronously.
+    
+    Args:
+        path: Path to JSON file
+        
+    Returns:
+        Parsed JSON data
+        
+    Raises:
+        RuntimeError: If file cannot be read or parsed
+    """
+    try:
+        async with aiofiles.open(path, 'r', encoding='utf-8') as f:
+            content = await f.read()
+            return json.loads(content)
+    except Exception as e:
+        raise RuntimeError(f"Failed to load JSON file {path}: {e}")
+
+
+async def load_model_weights(path: str, device: str = "cpu") -> dict:
+    """Load model weights asynchronously.
+    
+    Args:
+        path: Path to model file (.pth or .onnx)
+        device: Device to load model to
+        
+    Returns:
+        Model weights
+        
+    Raises:
+        RuntimeError: If file cannot be read
+    """
+    try:
+        async with aiofiles.open(path, 'rb') as f:
+            data = await f.read()
+            return torch.load(
+                io.BytesIO(data),
+                map_location=device,
+                weights_only=True
+            )
+    except Exception as e:
+        raise RuntimeError(f"Failed to load model weights from {path}: {e}")
+
+
 async def read_file(path: str) -> str:
     """Read text file asynchronously.
     
diff --git a/api/src/inference/__init__.py b/api/src/inference/__init__.py
index c98b884..98dc42c 100644
--- a/api/src/inference/__init__.py
+++ b/api/src/inference/__init__.py
@@ -1,4 +1,4 @@
-"""Inference backends and model management."""
+"""Model inference package."""
 
 from .base import BaseModelBackend
 from .model_manager import ModelManager, get_manager
@@ -6,15 +6,13 @@ from .onnx_cpu import ONNXCPUBackend
 from .onnx_gpu import ONNXGPUBackend
 from .pytorch_cpu import PyTorchCPUBackend
 from .pytorch_gpu import PyTorchGPUBackend
-from ..structures.model_schemas import ModelConfig
 
 __all__ = [
     'BaseModelBackend',
     'ModelManager',
     'get_manager',
-    'ModelConfig',
     'ONNXCPUBackend',
-    'ONNXGPUBackend', 
+    'ONNXGPUBackend',
     'PyTorchCPUBackend',
-    'PyTorchGPUBackend'
+    'PyTorchGPUBackend',
 ]
\ No newline at end of file
diff --git a/api/src/inference/model_manager.py b/api/src/inference/model_manager.py
index c3b76c7..26621c2 100644
--- a/api/src/inference/model_manager.py
+++ b/api/src/inference/model_manager.py
@@ -1,21 +1,18 @@
 """Model management and caching."""
 
-import os
-from typing import Dict, List, Optional, Union
+from typing import Dict, Optional
 
 import torch
 from loguru import logger
-from pydantic import BaseModel
 
+from ..core import paths
+from ..core.config import settings
+from ..core.model_config import ModelConfig, model_config
 from .base import BaseModelBackend
-from .voice_manager import get_manager as get_voice_manager
 from .onnx_cpu import ONNXCPUBackend
 from .onnx_gpu import ONNXGPUBackend
 from .pytorch_cpu import PyTorchCPUBackend
 from .pytorch_gpu import PyTorchGPUBackend
-from ..core import paths
-from ..core.config import settings
-from ..structures.model_schemas import ModelConfig
 
 
 class ModelManager:
@@ -27,44 +24,63 @@ class ModelManager:
         Args:
             config: Optional configuration
         """
-        self._config = config or ModelConfig()
+        self._config = config or model_config
         self._backends: Dict[str, BaseModelBackend] = {}
         self._current_backend: Optional[str] = None
-        self._voice_manager = get_voice_manager()
         self._initialize_backends()
 
     def _initialize_backends(self) -> None:
-        """Initialize available backends."""
-        """Initialize available backends."""
-        # Initialize GPU backends if available
-        if settings.use_gpu and torch.cuda.is_available():
-            try:
-                # PyTorch GPU
-                self._backends['pytorch_gpu'] = PyTorchGPUBackend()
-                self._current_backend = 'pytorch_gpu'
-                logger.info("Initialized PyTorch GPU backend")
-                
-                # ONNX GPU
-                self._backends['onnx_gpu'] = ONNXGPUBackend()
-                logger.info("Initialized ONNX GPU backend")
-            except Exception as e:
-                logger.error(f"Failed to initialize GPU backends: {e}")
-                # Fallback to CPU if GPU fails
+        """Initialize available backends based on settings."""
+        has_gpu = settings.use_gpu and torch.cuda.is_available()
+        
+        try:
+            if has_gpu:
+                if settings.use_onnx:
+                    # ONNX GPU primary
+                    self._backends['onnx_gpu'] = ONNXGPUBackend()
+                    self._current_backend = 'onnx_gpu'
+                    logger.info("Initialized ONNX GPU backend")
+                    
+                    # PyTorch GPU fallback
+                    self._backends['pytorch_gpu'] = PyTorchGPUBackend()
+                    logger.info("Initialized PyTorch GPU backend")
+                else:
+                    # PyTorch GPU primary
+                    self._backends['pytorch_gpu'] = PyTorchGPUBackend()
+                    self._current_backend = 'pytorch_gpu'
+                    logger.info("Initialized PyTorch GPU backend")
+                    
+                    # ONNX GPU fallback
+                    self._backends['onnx_gpu'] = ONNXGPUBackend()
+                    logger.info("Initialized ONNX GPU backend")
+            else:
                 self._initialize_cpu_backends()
-        else:
+        except Exception as e:
+            logger.error(f"Failed to initialize GPU backends: {e}")
+            # Fallback to CPU if GPU fails
             self._initialize_cpu_backends()
 
     def _initialize_cpu_backends(self) -> None:
-        """Initialize CPU backends."""
+        """Initialize CPU backends based on settings."""
         try:
-            # PyTorch CPU
-            self._backends['pytorch_cpu'] = PyTorchCPUBackend()
-            self._current_backend = 'pytorch_cpu'
-            logger.info("Initialized PyTorch CPU backend")
-            
-            # ONNX CPU
-            self._backends['onnx_cpu'] = ONNXCPUBackend()
-            logger.info("Initialized ONNX CPU backend")
+            if settings.use_onnx:
+                # ONNX CPU primary
+                self._backends['onnx_cpu'] = ONNXCPUBackend()
+                self._current_backend = 'onnx_cpu'
+                logger.info("Initialized ONNX CPU backend")
+                
+                # PyTorch CPU fallback
+                self._backends['pytorch_cpu'] = PyTorchCPUBackend()
+                logger.info("Initialized PyTorch CPU backend")
+            else:
+                # PyTorch CPU primary
+                self._backends['pytorch_cpu'] = PyTorchCPUBackend()
+                self._current_backend = 'pytorch_cpu'
+                logger.info("Initialized PyTorch CPU backend")
+                
+                # ONNX CPU fallback
+                self._backends['onnx_cpu'] = ONNXCPUBackend()
+                logger.info("Initialized ONNX CPU backend")
         except Exception as e:
             logger.error(f"Failed to initialize CPU backends: {e}")
             raise RuntimeError("No backends available")
@@ -98,7 +114,7 @@ class ModelManager:
         return self._backends[backend_type]
 
     def _determine_backend(self, model_path: str) -> str:
-        """Determine appropriate backend based on model file.
+        """Determine appropriate backend based on model file and settings.
         
         Args:
             model_path: Path to model file
@@ -106,10 +122,10 @@ class ModelManager:
         Returns:
             Backend type to use
         """
-        is_onnx = model_path.lower().endswith('.onnx')
         has_gpu = settings.use_gpu and torch.cuda.is_available()
         
-        if is_onnx:
+        # If ONNX is preferred or model is ONNX format
+        if settings.use_onnx or model_path.lower().endswith('.onnx'):
             return 'onnx_gpu' if has_gpu else 'onnx_cpu'
         else:
             return 'pytorch_gpu' if has_gpu else 'pytorch_cpu'
@@ -117,12 +133,14 @@ class ModelManager:
     async def load_model(
         self,
         model_path: str,
+        warmup_voice: Optional[torch.Tensor] = None,
         backend_type: Optional[str] = None
     ) -> None:
         """Load model on specified backend.
         
         Args:
             model_path: Path to model file
+            warmup_voice: Optional voice tensor for warmup, skips warmup if None
             backend_type: Backend to load on, uses default if None
             
         Raises:
@@ -138,35 +156,39 @@ class ModelManager:
             
             backend = self.get_backend(backend_type)
             
-            # Load model and run warmup
+            # Load model
             await backend.load_model(abs_path)
             logger.info(f"Loaded model on {backend_type} backend")
-            await self._warmup_inference(backend)
+            
+            # Run warmup if voice provided
+            if warmup_voice is not None:
+                await self._warmup_inference(backend, warmup_voice)
             
         except Exception as e:
             raise RuntimeError(f"Failed to load model: {e}")
             
-    async def _warmup_inference(self, backend: BaseModelBackend) -> None:
-        """Run warmup inference to initialize model."""
+    async def _warmup_inference(self, backend: BaseModelBackend, voice: torch.Tensor) -> None:
+        """Run warmup inference to initialize model.
+        
+        Args:
+            backend: Model backend to warm up
+            voice: Voice tensor already loaded on correct device
+        """
         try:
             # Import here to avoid circular imports
-            from ..text_processing import process_text
-            
-            # Load default voice for warmup
-            voice = await self._voice_manager.load_voice(settings.default_voice, device=backend.device)
-            logger.info(f"Loaded voice {settings.default_voice} for warmup")
+            from ..services.text_processing import process_text
             
             # Use real text
             text = "Testing text to speech synthesis."
-            logger.info(f"Running warmup inference with voice: af")
             
             # Process through pipeline
-            sequences = process_text(text)
-            if not sequences:
+            tokens = process_text(text)
+            if not tokens:
                 raise ValueError("Text processing failed")
             
             # Run inference
-            backend.generate(sequences[0], voice, speed=1.0)
+            backend.generate(tokens, voice, speed=1.0)
+            logger.info("Completed warmup inference")
             
         except Exception as e:
             logger.warning(f"Warmup inference failed: {e}")
@@ -175,7 +197,7 @@ class ModelManager:
     async def generate(
         self,
         tokens: list[int],
-        voice_name: str,
+        voice: torch.Tensor,
         speed: float = 1.0,
         backend_type: Optional[str] = None
     ) -> torch.Tensor:
@@ -183,7 +205,7 @@ class ModelManager:
         
         Args:
             tokens: Input token IDs
-            voice_name: Name of voice to use
+            voice: Voice tensor already loaded on correct device
             speed: Speed multiplier
             backend_type: Backend to use, uses default if None
             
@@ -198,10 +220,7 @@ class ModelManager:
             raise RuntimeError("Model not loaded")
 
         try:
-            # Load voice using voice manager
-            voice = await self._voice_manager.load_voice(voice_name, device=backend.device)
-            
-            # Generate audio
+            # Generate audio using provided voice tensor
             return backend.generate(tokens, voice, speed)
             
         except Exception as e:
diff --git a/api/src/inference/onnx_cpu.py b/api/src/inference/onnx_cpu.py
index 600157f..34d68ac 100644
--- a/api/src/inference/onnx_cpu.py
+++ b/api/src/inference/onnx_cpu.py
@@ -13,8 +13,7 @@ from onnxruntime import (
 )
 
 from ..core import paths
-from ..core.config import settings
-from ..structures.model_schemas import ONNXConfig
+from ..core.model_config import model_config
 from .base import BaseModelBackend
 
 
@@ -26,14 +25,11 @@ class ONNXCPUBackend(BaseModelBackend):
         super().__init__()
         self._device = "cpu"
         self._session: Optional[InferenceSession] = None
-        self._config = ONNXConfig(
-            optimization_level=settings.onnx_optimization_level,
-            num_threads=settings.onnx_num_threads,
-            inter_op_threads=settings.onnx_inter_op_threads,
-            execution_mode=settings.onnx_execution_mode,
-            memory_pattern=settings.onnx_memory_pattern,
-            arena_extend_strategy=settings.onnx_arena_extend_strategy
-        )
+
+    @property
+    def is_loaded(self) -> bool:
+        """Check if model is loaded."""
+        return self._session is not None
 
     async def load_model(self, path: str) -> None:
         """Load ONNX model.
@@ -115,28 +111,29 @@ class ONNXCPUBackend(BaseModelBackend):
             Configured session options
         """
         options = SessionOptions()
+        config = model_config.onnx_cpu
         
         # Set optimization level
-        if self._config.optimization_level == "all":
+        if config.optimization_level == "all":
             options.graph_optimization_level = GraphOptimizationLevel.ORT_ENABLE_ALL
-        elif self._config.optimization_level == "basic":
+        elif config.optimization_level == "basic":
             options.graph_optimization_level = GraphOptimizationLevel.ORT_ENABLE_BASIC
         else:
             options.graph_optimization_level = GraphOptimizationLevel.ORT_DISABLE_ALL
         
         # Configure threading
-        options.intra_op_num_threads = self._config.num_threads
-        options.inter_op_num_threads = self._config.inter_op_threads
+        options.intra_op_num_threads = config.num_threads
+        options.inter_op_num_threads = config.inter_op_threads
         
         # Set execution mode
         options.execution_mode = (
             ExecutionMode.ORT_PARALLEL
-            if self._config.execution_mode == "parallel"
+            if config.execution_mode == "parallel"
             else ExecutionMode.ORT_SEQUENTIAL
         )
         
         # Configure memory optimization
-        options.enable_mem_pattern = self._config.memory_pattern
+        options.enable_mem_pattern = config.memory_pattern
         
         return options
 
@@ -148,7 +145,15 @@ class ONNXCPUBackend(BaseModelBackend):
         """
         return {
             "CPUExecutionProvider": {
-                "arena_extend_strategy": self._config.arena_extend_strategy,
+                "arena_extend_strategy": model_config.onnx_cpu.arena_extend_strategy,
                 "cpu_memory_arena_cfg": "cpu:0"
             }
-        }
\ No newline at end of file
+        }
+
+    def unload(self) -> None:
+        """Unload model and free resources."""
+        if self._session is not None:
+            del self._session
+            self._session = None
+            if torch.cuda.is_available():
+                torch.cuda.empty_cache()
\ No newline at end of file
diff --git a/api/src/inference/onnx_gpu.py b/api/src/inference/onnx_gpu.py
index d003a86..2df32e2 100644
--- a/api/src/inference/onnx_gpu.py
+++ b/api/src/inference/onnx_gpu.py
@@ -13,8 +13,7 @@ from onnxruntime import (
 )
 
 from ..core import paths
-from ..core.config import settings
-from ..structures.model_schemas import ONNXGPUConfig
+from ..core.model_config import model_config
 from .base import BaseModelBackend
 
 
@@ -28,18 +27,11 @@ class ONNXGPUBackend(BaseModelBackend):
             raise RuntimeError("CUDA not available")
         self._device = "cuda"
         self._session: Optional[InferenceSession] = None
-        self._config = ONNXGPUConfig(
-            optimization_level=settings.onnx_optimization_level,
-            num_threads=settings.onnx_num_threads,
-            inter_op_threads=settings.onnx_inter_op_threads,
-            execution_mode=settings.onnx_execution_mode,
-            memory_pattern=settings.onnx_memory_pattern,
-            arena_extend_strategy=settings.onnx_arena_extend_strategy,
-            device_id=0,
-            gpu_mem_limit=0.7,
-            cudnn_conv_algo_search="EXHAUSTIVE",
-            do_copy_in_default_stream=True
-        )
+
+    @property
+    def is_loaded(self) -> bool:
+        """Check if model is loaded."""
+        return self._session is not None
 
     async def load_model(self, path: str) -> None:
         """Load ONNX model.
@@ -121,28 +113,29 @@ class ONNXGPUBackend(BaseModelBackend):
             Configured session options
         """
         options = SessionOptions()
+        config = model_config.onnx_gpu
         
         # Set optimization level
-        if self._config.optimization_level == "all":
+        if config.optimization_level == "all":
             options.graph_optimization_level = GraphOptimizationLevel.ORT_ENABLE_ALL
-        elif self._config.optimization_level == "basic":
+        elif config.optimization_level == "basic":
             options.graph_optimization_level = GraphOptimizationLevel.ORT_ENABLE_BASIC
         else:
             options.graph_optimization_level = GraphOptimizationLevel.ORT_DISABLE_ALL
         
         # Configure threading
-        options.intra_op_num_threads = self._config.num_threads
-        options.inter_op_num_threads = self._config.inter_op_threads
+        options.intra_op_num_threads = config.num_threads
+        options.inter_op_num_threads = config.inter_op_threads
         
         # Set execution mode
         options.execution_mode = (
             ExecutionMode.ORT_PARALLEL
-            if self._config.execution_mode == "parallel"
+            if config.execution_mode == "parallel"
             else ExecutionMode.ORT_SEQUENTIAL
         )
         
         # Configure memory optimization
-        options.enable_mem_pattern = self._config.memory_pattern
+        options.enable_mem_pattern = config.memory_pattern
         
         return options
 
@@ -152,12 +145,21 @@ class ONNXGPUBackend(BaseModelBackend):
         Returns:
             Provider configuration
         """
+        config = model_config.onnx_gpu
         return {
             "CUDAExecutionProvider": {
-                "device_id": self._config.device_id,
-                "arena_extend_strategy": self._config.arena_extend_strategy,
-                "gpu_mem_limit": int(self._config.gpu_mem_limit * torch.cuda.get_device_properties(0).total_memory),
-                "cudnn_conv_algo_search": self._config.cudnn_conv_algo_search,
-                "do_copy_in_default_stream": self._config.do_copy_in_default_stream
+                "device_id": config.device_id,
+                "arena_extend_strategy": config.arena_extend_strategy,
+                "gpu_mem_limit": int(config.gpu_mem_limit * torch.cuda.get_device_properties(0).total_memory),
+                "cudnn_conv_algo_search": config.cudnn_conv_algo_search,
+                "do_copy_in_default_stream": config.do_copy_in_default_stream
             }
-        }
\ No newline at end of file
+        }
+
+    def unload(self) -> None:
+        """Unload model and free resources."""
+        if self._session is not None:
+            del self._session
+            self._session = None
+            if torch.cuda.is_available():
+                torch.cuda.empty_cache()
\ No newline at end of file
diff --git a/api/src/inference/pytorch_cpu.py b/api/src/inference/pytorch_cpu.py
index e3fe5fa..b1d4328 100644
--- a/api/src/inference/pytorch_cpu.py
+++ b/api/src/inference/pytorch_cpu.py
@@ -9,7 +9,7 @@ from loguru import logger
 
 from ..builds.models import build_model
 from ..core import paths
-from ..structures.model_schemas import PyTorchCPUConfig
+from ..core.model_config import model_config
 from .base import BaseModelBackend
 
 
@@ -118,12 +118,12 @@ class PyTorchCPUBackend(BaseModelBackend):
         super().__init__()
         self._device = "cpu"
         self._model: Optional[torch.nn.Module] = None
-        self._config = PyTorchCPUConfig()
 
         # Configure PyTorch CPU settings
-        if self._config.num_threads > 0:
-            torch.set_num_threads(self._config.num_threads)
-        if self._config.pin_memory:
+        config = model_config.pytorch_cpu
+        if config.num_threads > 0:
+            torch.set_num_threads(config.num_threads)
+        if config.pin_memory:
             torch.set_default_tensor_type(torch.FloatTensor)
 
     async def load_model(self, path: str) -> None:
diff --git a/api/src/inference/pytorch_gpu.py b/api/src/inference/pytorch_gpu.py
index 3b41f9f..2995818 100644
--- a/api/src/inference/pytorch_gpu.py
+++ b/api/src/inference/pytorch_gpu.py
@@ -9,7 +9,7 @@ from loguru import logger
 
 from ..builds.models import build_model
 from ..core import paths
-from ..structures.model_schemas import PyTorchConfig
+from ..core.model_config import model_config
 from .base import BaseModelBackend
 
 
@@ -96,7 +96,12 @@ class PyTorchGPUBackend(BaseModelBackend):
             raise RuntimeError("CUDA not available")
         self._device = "cuda"
         self._model: Optional[torch.nn.Module] = None
-        self._config = PyTorchConfig()
+        
+        # Configure GPU settings
+        config = model_config.pytorch_gpu
+        if config.sync_cuda:
+            torch.cuda.synchronize()
+        torch.cuda.set_device(config.device_id)
 
     async def load_model(self, path: str) -> None:
         """Load PyTorch model.
@@ -154,13 +159,19 @@ class PyTorchGPUBackend(BaseModelBackend):
             
         except Exception as e:
             logger.error(f"Generation failed: {e}")
+            if model_config.pytorch_gpu.retry_on_oom and "out of memory" in str(e).lower():
+                self._clear_memory()
+                return self.generate(tokens, voice, speed)  # Retry once
             raise
+        finally:
+            if model_config.pytorch_gpu.sync_cuda:
+                torch.cuda.synchronize()
 
     def _check_memory(self) -> bool:
         """Check if memory usage is above threshold."""
         if torch.cuda.is_available():
             memory_gb = torch.cuda.memory_allocated() / 1e9
-            return memory_gb > self._config.memory_threshold
+            return memory_gb > model_config.pytorch_gpu.memory_threshold
         return False
 
     def _clear_memory(self) -> None:
diff --git a/api/src/inference/voice_manager.py b/api/src/inference/voice_manager.py
index 295040c..15932c7 100644
--- a/api/src/inference/voice_manager.py
+++ b/api/src/inference/voice_manager.py
@@ -33,7 +33,15 @@ class VoiceManager:
         Returns:
             Path to voice file if exists, None otherwise
         """
-        voice_path = os.path.join(settings.voices_dir, f"{voice_name}.pt")
+        # Get api directory path (two levels up from inference)
+        api_dir = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
+        
+        # Construct voice path relative to api directory
+        voice_path = os.path.join(api_dir, settings.voices_dir, f"{voice_name}.pt")
+        
+        # Ensure voices directory exists
+        os.makedirs(os.path.dirname(voice_path), exist_ok=True)
+        
         return voice_path if os.path.exists(voice_path) else None
 
     async def load_voice(self, voice_name: str, device: str = "cpu") -> torch.Tensor:
@@ -112,8 +120,15 @@ class VoiceManager:
             combined_name = "_".join(voices)
             combined_tensor = torch.mean(torch.stack(voice_tensors), dim=0)
             
+            # Get api directory path
+            api_dir = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
+            voices_dir = os.path.join(api_dir, settings.voices_dir)
+            
+            # Ensure voices directory exists
+            os.makedirs(voices_dir, exist_ok=True)
+            
             # Save combined voice
-            combined_path = os.path.join(settings.voices_dir, f"{combined_name}.pt")
+            combined_path = os.path.join(voices_dir, f"{combined_name}.pt")
             try:
                 torch.save(combined_tensor, combined_path)
             except Exception as e:
@@ -132,7 +147,15 @@ class VoiceManager:
         """
         voices = []
         try:
-            for entry in os.listdir(settings.voices_dir):
+            # Get api directory path
+            api_dir = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
+            voices_dir = os.path.join(api_dir, settings.voices_dir)
+            
+            # Ensure voices directory exists
+            os.makedirs(voices_dir, exist_ok=True)
+            
+            # List voice files
+            for entry in os.listdir(voices_dir):
                 if entry.endswith(".pt"):
                     voices.append(entry[:-3])  # Remove .pt extension
         except Exception as e:
diff --git a/api/src/main.py b/api/src/main.py
index 4effda2..ec1b6c8 100644
--- a/api/src/main.py
+++ b/api/src/main.py
@@ -13,7 +13,6 @@ from loguru import logger
 from .core.config import settings
 from .routers.development import router as dev_router
 from .routers.openai_compatible import router as openai_router
-from .services.tts_model import TTSModel
 from .services.tts_service import TTSService
 
 
@@ -44,25 +43,32 @@ async def lifespan(app: FastAPI):
     """Lifespan context manager for model initialization"""
     logger.info("Loading TTS model and voice packs...")
 
-    # Initialize the main model with warm-up
-    voicepack_count = await TTSModel.setup()
-    # boundary = "█████╗"*9
+    # Initialize service
+    service = TTSService()
+    await service.ensure_initialized()
+    
+    # Get available voices
+    voices = await service.list_voices()
+    voicepack_count = len(voices)
+
+    # Get device info from model manager
+    device = "GPU" if settings.use_gpu else "CPU"
+    model = "ONNX" if settings.use_onnx else "PyTorch"
     boundary = "░" * 2*12
     startup_msg = f"""
 
 {boundary}
 
     ╔═╗┌─┐┌─┐┌┬┐
-    ╠╣ ├─┤└─┐ │ 
-    ╚  ┴ ┴└─┘ ┴ 
+    ╠╣ ├─┤└─┐ │
+    ╚  ┴ ┴└─┘ ┴
     ╦╔═┌─┐┬┌─┌─┐
     ╠╩╗│ │├┴┐│ │
     ╩ ╩└─┘┴ ┴└─┘
 
 {boundary}
                 """
-    # TODO: Improve CPU warmup, threads, memory, etc
-    startup_msg += f"\nModel warmed up on {TTSModel.get_device()}"
+    startup_msg += f"\nModel warmed up on {device}: {model}"
     startup_msg += f"\n{voicepack_count} voice packs loaded\n"
     startup_msg += f"\n{boundary}\n"
     logger.info(startup_msg)
diff --git a/api/src/routers/development.py b/api/src/routers/development.py
index bc43eb0..4713440 100644
--- a/api/src/routers/development.py
+++ b/api/src/routers/development.py
@@ -6,7 +6,6 @@ from loguru import logger
 
 from ..services.audio import AudioService
 from ..services.text_processing import phonemize, tokenize
-from ..services.tts_model import TTSModel
 from ..services.tts_service import TTSService
 from ..structures.text_schemas import (
     GenerateFromPhonemesRequest,
@@ -82,27 +81,34 @@ async def generate_from_phonemes(
             detail={"error": "Invalid request", "message": "Phonemes cannot be empty"},
         )
 
-    # Validate voice exists
-    voice_path = tts_service._get_voice_path(request.voice)
-    if not voice_path:
-        raise HTTPException(
-            status_code=400,
-            detail={
-                "error": "Invalid request",
-                "message": f"Voice not found: {request.voice}",
-            },
-        )
-
     try:
-        # Load voice
-        voicepack = tts_service._load_voice(voice_path)
+        # Ensure service is initialized
+        await tts_service.ensure_initialized()
+
+        # Validate voice exists
+        available_voices = await tts_service.list_voices()
+        if request.voice not in available_voices:
+            raise HTTPException(
+                status_code=400,
+                detail={
+                    "error": "Invalid request",
+                    "message": f"Voice not found: {request.voice}",
+                },
+            )
 
         # Convert phonemes to tokens
         tokens = tokenize(request.phonemes)
         tokens = [0] + tokens + [0]  # Add start/end tokens
 
         # Generate audio directly from tokens
-        audio = TTSModel.generate_from_tokens(tokens, voicepack, request.speed)
+        audio = await tts_service.model_manager.generate(
+            tokens,
+            request.voice,
+            speed=request.speed
+        )
+
+        if audio is None:
+            raise ValueError("Failed to generate audio")
 
         # Convert to WAV bytes
         wav_bytes = AudioService.convert_audio(
diff --git a/api/src/services/text_processing/__init__.py b/api/src/services/text_processing/__init__.py
index 8427232..c9b3eb4 100644
--- a/api/src/services/text_processing/__init__.py
+++ b/api/src/services/text_processing/__init__.py
@@ -1,13 +1,28 @@
-from .normalizer import normalize_text
-from .phonemizer import EspeakBackend, PhonemizerBackend, phonemize
-from .vocabulary import VOCAB, decode_tokens, tokenize
+"""Text processing pipeline."""
 
-__all__ = [
-    "normalize_text",
-    "phonemize",
-    "tokenize",
-    "decode_tokens",
-    "VOCAB",
-    "PhonemizerBackend",
-    "EspeakBackend",
-]
+from .chunker import split_text
+from .normalizer import normalize_text
+from .phonemizer import phonemize
+from .vocabulary import tokenize
+
+
+def process_text(text: str, language: str = "a") -> list[int]:
+    """Process text through the full pipeline.
+    
+    Args:
+        text: Input text
+        language: Language code ('a' for US English, 'b' for British English)
+        
+    Returns:
+        List of token IDs
+        
+    Note:
+        The pipeline:
+        1. Converts text to phonemes using phonemizer
+        2. Converts phonemes to token IDs using vocabulary
+    """
+    # Convert text to phonemes
+    phonemes = phonemize(text, language=language)
+    
+    # Convert phonemes to token IDs
+    return tokenize(phonemes)
diff --git a/api/src/services/tts_service.py b/api/src/services/tts_service.py
index 20e9b69..9c51b19 100644
--- a/api/src/services/tts_service.py
+++ b/api/src/services/tts_service.py
@@ -5,17 +5,16 @@ import os
 import time
 from typing import List, Tuple
 
-import torch
-
 import numpy as np
 import scipy.io.wavfile as wavfile
+import torch
 from loguru import logger
 
 from ..core.config import settings
 from ..inference.model_manager import get_manager as get_model_manager
 from ..inference.voice_manager import get_manager as get_voice_manager
 from .audio import AudioNormalizer, AudioService
-from .text_processing import chunker, normalize_text
+from .text_processing import chunker, normalize_text, process_text
 
 
 class TTSService:
@@ -41,16 +40,33 @@ class TTSService:
             raise self._initialization_error
 
         try:
-            # Determine model path based on hardware
+            # Get api directory path (one level up from src)
+            api_dir = os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
+            
+            # Determine model file and backend based on hardware
             if settings.use_gpu and torch.cuda.is_available():
-                model_path = os.path.join(settings.model_dir, settings.pytorch_model_path)
+                model_file = settings.pytorch_model_file
                 backend_type = 'pytorch_gpu'
             else:
-                model_path = os.path.join(settings.model_dir, settings.onnx_model_path)
+                model_file = settings.onnx_model_file
                 backend_type = 'onnx_cpu'
             
-            # Initialize model
-            await self.model_manager.load_model(model_path, backend_type)
+            # Construct model path relative to api directory
+            model_path = os.path.join(api_dir, settings.model_dir, model_file)
+            
+            # Ensure model directory exists
+            os.makedirs(os.path.dirname(model_path), exist_ok=True)
+            
+            if not os.path.exists(model_path):
+                raise RuntimeError(f"Model file not found: {model_path}")
+            
+            # Load default voice for warmup
+            backend = self.model_manager.get_backend(backend_type)
+            warmup_voice = await self.voice_manager.load_voice(settings.default_voice, device=backend.device)
+            logger.info(f"Loaded voice {settings.default_voice} for warmup")
+            
+            # Initialize model with warmup voice
+            await self.model_manager.load_model(model_path, warmup_voice, backend_type)
             logger.info(f"Initialized model on {backend_type} backend")
             self._initialized = True
             
@@ -86,16 +102,19 @@ class TTSService:
             audio_chunks = []
             for chunk in chunker.split_text(text):
                 try:
-                    # Process text
-                    
-                    sequences = process_text(chunk)
-                    if not sequences:
+                    # Convert chunk to token IDs
+                    tokens = process_text(chunk)
+                    if not tokens:
                         continue
 
+                    # Get backend and load voice
+                    backend = self.model_manager.get_backend()
+                    voice_tensor = await self.voice_manager.load_voice(voice, device=backend.device)
+                    
                     # Generate audio
                     chunk_audio = await self.model_manager.generate(
-                        sequences[0],
-                        voice,
+                        tokens,
+                        voice_tensor,
                         speed=speed
                     )
                     if chunk_audio is not None:
@@ -154,14 +173,17 @@ class TTSService:
             while current_chunk is not None:
                 next_chunk = next(chunk_gen, None)
                 try:
-                    # Process text
-                    from ..text_processing import process_text
-                    sequences = process_text(current_chunk)
-                    if sequences:
+                    # Convert chunk to token IDs
+                    tokens = process_text(current_chunk)
+                    if tokens:
+                        # Get backend and load voice
+                        backend = self.model_manager.get_backend()
+                        voice_tensor = await self.voice_manager.load_voice(voice, device=backend.device)
+                        
                         # Generate audio
                         chunk_audio = await self.model_manager.generate(
-                            sequences[0],
-                            voice,
+                            tokens,
+                            voice_tensor,
                             speed=speed
                         )
 
diff --git a/api/src/structures/model_schemas.py b/api/src/structures/model_schemas.py
index 51ef910..b6ed61d 100644
--- a/api/src/structures/model_schemas.py
+++ b/api/src/structures/model_schemas.py
@@ -1,26 +1,13 @@
-"""Model and voice configuration schemas."""
+"""Voice configuration schemas."""
 
-from pydantic import BaseModel
-
-
-class ModelConfig(BaseModel):
-    """Model configuration."""
-    optimization_level: str = "all"  # all, basic, none
-    num_threads: int = 4
-    inter_op_threads: int = 4
-    execution_mode: str = "parallel"  # parallel, sequential
-    memory_pattern: bool = True
-    arena_extend_strategy: str = "kNextPowerOfTwo"
-
-    class Config:
-        frozen = True  # Make config immutable
+from pydantic import BaseModel, Field
 
 
 class VoiceConfig(BaseModel):
     """Voice configuration."""
-    use_cache: bool = True
-    cache_size: int = 3  # Number of voices to cache
-    validate_on_load: bool = True  # Whether to validate voices when loading
+    use_cache: bool = Field(True, description="Whether to cache loaded voices")
+    cache_size: int = Field(3, description="Number of voices to cache")
+    validate_on_load: bool = Field(True, description="Whether to validate voices when loading")
 
     class Config:
         frozen = True  # Make config immutable
\ No newline at end of file
diff --git a/docker/cpu/Dockerfile b/docker/cpu/Dockerfile
index e4bd32c..0613022 100644
--- a/docker/cpu/Dockerfile
+++ b/docker/cpu/Dockerfile
@@ -17,29 +17,10 @@ COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 RUN useradd -m -u 1000 appuser
 
 # Create directories and set ownership
-RUN mkdir -p /app/models && \
-    mkdir -p /app/api/src/voices && \
+RUN mkdir -p /app/api/src/voices && \
     chown -R appuser:appuser /app
 
 USER appuser
-
-# Download and extract models
-WORKDIR /app/models
-RUN set -x && \
-    curl -L -o model.tar.gz https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.0.1/kokoro-82m-onnx.tar.gz && \
-    echo "Downloaded model.tar.gz:" && ls -lh model.tar.gz && \
-    tar xzf model.tar.gz && \
-    echo "Contents after extraction:" && ls -lhR && \
-    rm model.tar.gz && \
-    echo "Final contents:" && ls -lhR
-
-# Download and extract voice models
-WORKDIR /app/api/src/voices
-RUN curl -L -o voices.tar.gz https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.0.1/voice-models.tar.gz && \
-    tar xzf voices.tar.gz && \
-    rm voices.tar.gz
-
-# Switch back to app directory
 WORKDIR /app
 
 # Copy dependency files
@@ -59,9 +40,10 @@ RUN --mount=type=cache,target=/root/.cache/uv \
 
 # Set environment variables
 ENV PYTHONUNBUFFERED=1
-ENV PYTHONPATH=/app:/app/models
+ENV PYTHONPATH=/app
 ENV PATH="/app/.venv/bin:$PATH"
 ENV UV_LINK_MODE=copy
+ENV USE_GPU=false
 
 # Run FastAPI server
 CMD ["uv", "run", "python", "-m", "uvicorn", "api.src.main:app", "--host", "0.0.0.0", "--port", "8880", "--log-level", "debug"]
diff --git a/docker/cpu/docker-compose.yml b/docker/cpu/docker-compose.yml
index 0f9003a..2138594 100644
--- a/docker/cpu/docker-compose.yml
+++ b/docker/cpu/docker-compose.yml
@@ -6,12 +6,11 @@ services:
       context: ../..
       dockerfile: docker/cpu/Dockerfile
     volumes:
-      - ../../api/src:/app/api/src
-      - ../../api/src/voices:/app/api/src/voices
+      - ../../api:/app/api
     ports:
       - "8880:8880"
     environment:
-      - PYTHONPATH=/app:/app/models
+      - PYTHONPATH=/app:/app/api
       # ONNX Optimization Settings for vectorized operations
       - ONNX_NUM_THREADS=8  # Maximize core usage for vectorized ops
       - ONNX_INTER_OP_THREADS=4  # Higher inter-op for parallel matrix operations
@@ -20,20 +19,20 @@ services:
       - ONNX_MEMORY_PATTERN=true
       - ONNX_ARENA_EXTEND_STRATEGY=kNextPowerOfTwo
       
-  # Gradio UI service [Comment out everything below if you don't need it]
-  gradio-ui:
-    image: ghcr.io/remsky/kokoro-fastapi-ui:v0.1.0
-    # Uncomment below (and comment out above) to build from source instead of using the released image
-    build:
-      context: ../../ui
-    ports:
-      - "7860:7860"
-    volumes:
-      - ../../ui/data:/app/ui/data
-      - ../../ui/app.py:/app/app.py  # Mount app.py for hot reload
-    environment:
-      - GRADIO_WATCH=True  # Enable hot reloading
-      - PYTHONUNBUFFERED=1  # Ensure Python output is not buffered
-      - DISABLE_LOCAL_SAVING=false  # Set to 'true' to disable local saving and hide file view
-      - API_HOST=kokoro-tts  # Set TTS service URL
-      - API_PORT=8880  # Set TTS service PORT
+  # # Gradio UI service [Comment out everything below if you don't need it]
+  # gradio-ui:
+  #   image: ghcr.io/remsky/kokoro-fastapi-ui:v0.1.0
+  #   # Uncomment below (and comment out above) to build from source instead of using the released image
+  #   build:
+  #     context: ../../ui
+  #   ports:
+  #     - "7860:7860"
+  #   volumes:
+  #     - ../../ui/data:/app/ui/data
+  #     - ../../ui/app.py:/app/app.py  # Mount app.py for hot reload
+  #   environment:
+  #     - GRADIO_WATCH=True  # Enable hot reloading
+  #     - PYTHONUNBUFFERED=1  # Ensure Python output is not buffered
+  #     - DISABLE_LOCAL_SAVING=false  # Set to 'true' to disable local saving and hide file view
+  #     - API_HOST=kokoro-tts  # Set TTS service URL
+  #     - API_PORT=8880  # Set TTS service PORT
diff --git a/docker/cpu/download_onnx.py b/docker/cpu/download_onnx.py
new file mode 100755
index 0000000..a04718c
--- /dev/null
+++ b/docker/cpu/download_onnx.py
@@ -0,0 +1,53 @@
+#!/usr/bin/env python3
+import os
+import sys
+import requests
+from pathlib import Path
+from typing import List
+
+def download_file(url: str, output_dir: Path) -> None:
+    """Download a file from URL to the specified directory."""
+    filename = os.path.basename(url)
+    output_path = output_dir / filename
+    
+    print(f"Downloading {filename}...")
+    response = requests.get(url, stream=True)
+    response.raise_for_status()
+    
+    with open(output_path, 'wb') as f:
+        for chunk in response.iter_content(chunk_size=8192):
+            f.write(chunk)
+
+def find_project_root() -> Path:
+    """Find project root by looking for api directory."""
+    max_steps = 5
+    current = Path(__file__).resolve()
+    for _ in range(max_steps):
+        if (current / 'api').is_dir():
+            return current
+        current = current.parent
+    raise RuntimeError("Could not find project root (no api directory found)")
+
+def main(custom_models: List[str] = None):
+    # Always use top-level models directory relative to project root
+    project_root = find_project_root()
+    models_dir = project_root / 'api' / 'src' / 'models'
+    models_dir.mkdir(exist_ok=True, parents=True)
+    
+    # Default ONNX model if no arguments provided
+    default_models = [
+        "https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.1.0/kokoro-v0_19.onnx",
+        "https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.1.0/kokoro-v0_19_fp16.onnx"
+    ]
+    
+    # Use provided models or default
+    models_to_download = custom_models if custom_models else default_models
+    
+    for model_url in models_to_download:
+        try:
+            download_file(model_url, models_dir)
+        except Exception as e:
+            print(f"Error downloading {model_url}: {e}")
+
+if __name__ == "__main__":
+    main(sys.argv[1:] if len(sys.argv) > 1 else None)
\ No newline at end of file
diff --git a/docker/cpu/download_onnx.sh b/docker/cpu/download_onnx.sh
new file mode 100755
index 0000000..c0a250b
--- /dev/null
+++ b/docker/cpu/download_onnx.sh
@@ -0,0 +1,32 @@
+#!/bin/bash
+
+# Ensure models directory exists
+mkdir -p api/src/models
+
+# Function to download a file
+download_file() {
+    local url="$1"
+    local filename=$(basename "$url")
+    echo "Downloading $filename..."
+    curl -L "$url" -o "api/src/models/$filename"
+}
+
+# Default ONNX model if no arguments provided
+DEFAULT_MODELS=(
+    "https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.1.0/kokoro-v0_19.onnx"
+    "https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.1.0/kokoro-v0_19_fp16.onnx"
+)
+
+# Use provided models or default
+if [ $# -gt 0 ]; then
+    MODELS=("$@")
+else
+    MODELS=("${DEFAULT_MODELS[@]}")
+fi
+
+# Download all models
+for model in "${MODELS[@]}"; do
+    download_file "$model"
+done
+
+echo "ONNX model download complete!"
\ No newline at end of file
diff --git a/docker/gpu/Dockerfile b/docker/gpu/Dockerfile
index ed4676d..e6bf9ff 100644
--- a/docker/gpu/Dockerfile
+++ b/docker/gpu/Dockerfile
@@ -1,9 +1,8 @@
-FROM nvidia/cuda:12.1.0-base-ubuntu22.04
-
-# Install Python and other dependencies
+FROM pytorch/pytorch:2.1.0-cuda11.8-cudnn8-runtime
+# Set non-interactive frontend
+ENV DEBIAN_FRONTEND=noninteractive
+# Install dependencies
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    python3.10 \
-    python3.10-venv \
     espeak-ng \
     git \
     libsndfile1 \
@@ -19,25 +18,10 @@ COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 RUN useradd -m -u 1000 appuser
 
 # Create directories and set ownership
-RUN mkdir -p /app/models && \
-    mkdir -p /app/api/src/voices && \
+RUN mkdir -p /app/api/src/voices && \
     chown -R appuser:appuser /app
 
 USER appuser
-
-# Download and extract models
-WORKDIR /app/models
-RUN curl -L -o model.tar.gz https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.0.1/kokoro-82m-pytorch.tar.gz && \
-    tar xzf model.tar.gz && \
-    rm model.tar.gz
-
-# Download and extract voice models
-WORKDIR /app/api/src/voices
-RUN curl -L -o voices.tar.gz https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.0.1/voice-models.tar.gz && \
-    tar xzf voices.tar.gz && \
-    rm voices.tar.gz
-
-# Switch back to app directory
 WORKDIR /app
 
 # Copy dependency files
@@ -57,9 +41,10 @@ RUN --mount=type=cache,target=/root/.cache/uv \
 
 # Set environment variables
 ENV PYTHONUNBUFFERED=1
-ENV PYTHONPATH=/app:/app/models
+ENV PYTHONPATH=/app
 ENV PATH="/app/.venv/bin:$PATH"
 ENV UV_LINK_MODE=copy
+ENV USE_GPU=true
 
 # Run FastAPI server
 CMD ["uv", "run", "python", "-m", "uvicorn", "api.src.main:app", "--host", "0.0.0.0", "--port", "8880", "--log-level", "debug"]
diff --git a/docker/gpu/docker-compose.yml b/docker/gpu/docker-compose.yml
index 9433dc6..458eff9 100644
--- a/docker/gpu/docker-compose.yml
+++ b/docker/gpu/docker-compose.yml
@@ -6,12 +6,14 @@ services:
       context: ../..
       dockerfile: docker/gpu/Dockerfile
     volumes:
-      - ../../api/src:/app/api/src  # Mount src for development
-      - ../../api/src/voices:/app/api/src/voices  # Mount voices for persistence
+      - ../../api:/app/api
     ports:
       - "8880:8880"
     environment:
-      - PYTHONPATH=/app:/app/models
+      - PYTHONPATH=/app
+      - USE_GPU=true
+      - USE_ONNX=false
+      - PYTHONUNBUFFERED=1
     deploy:
       resources:
         reservations:
@@ -20,20 +22,20 @@ services:
               count: 1
               capabilities: [gpu]
 
-  # Gradio UI service
-  gradio-ui:
-    image: ghcr.io/remsky/kokoro-fastapi-ui:v0.1.0
-    # Uncomment below to build from source instead of using the released image
-    # build:
-      # context: ../../ui
-    ports:
-      - "7860:7860"
-    volumes:
-      - ../../ui/data:/app/ui/data
-      - ../../ui/app.py:/app/app.py  # Mount app.py for hot reload
-    environment:
-      - GRADIO_WATCH=1  # Enable hot reloading
-      - PYTHONUNBUFFERED=1  # Ensure Python output is not buffered
-      - DISABLE_LOCAL_SAVING=false  # Set to 'true' to disable local saving and hide file view
-      - API_HOST=kokoro-tts  # Set TTS service URL
-      - API_PORT=8880  # Set TTS service PORT
+  # # Gradio UI service
+  # gradio-ui:
+  #   image: ghcr.io/remsky/kokoro-fastapi-ui:v0.1.0
+  #   # Uncomment below to build from source instead of using the released image
+  #   # build:
+  #     # context: ../../ui
+  #   ports:
+  #     - "7860:7860"
+  #   volumes:
+  #     - ../../ui/data:/app/ui/data
+  #     - ../../ui/app.py:/app/app.py  # Mount app.py for hot reload
+  #   environment:
+  #     - GRADIO_WATCH=1  # Enable hot reloading
+  #     - PYTHONUNBUFFERED=1  # Ensure Python output is not buffered
+  #     - DISABLE_LOCAL_SAVING=false  # Set to 'true' to disable local saving and hide file view
+  #     - API_HOST=kokoro-tts  # Set TTS service URL
+  #     - API_PORT=8880  # Set TTS service PORT
diff --git a/docker/gpu/download_pth.py b/docker/gpu/download_pth.py
new file mode 100755
index 0000000..f58f29d
--- /dev/null
+++ b/docker/gpu/download_pth.py
@@ -0,0 +1,57 @@
+#!/usr/bin/env python3
+import os
+import sys
+import requests
+from pathlib import Path
+from typing import List
+
+def download_file(url: str, output_dir: Path) -> None:
+    """Download a file from URL to the specified directory."""
+    filename = os.path.basename(url)
+    if not filename.endswith('.pth'):
+        print(f"Warning: {filename} is not a .pth file")
+        return
+        
+    output_path = output_dir / filename
+    
+    print(f"Downloading {filename}...")
+    response = requests.get(url, stream=True)
+    response.raise_for_status()
+    
+    with open(output_path, 'wb') as f:
+        for chunk in response.iter_content(chunk_size=8192):
+            f.write(chunk)
+
+def find_project_root() -> Path:
+    """Find project root by looking for api directory."""
+    max_steps = 5
+    current = Path(__file__).resolve()
+    for _ in range(max_steps):
+        if (current / 'api').is_dir():
+            return current
+        current = current.parent
+    raise RuntimeError("Could not find project root (no api directory found)")
+
+def main(custom_models: List[str] = None):
+    # Find project root and ensure models directory exists
+    project_root = find_project_root()
+    models_dir = project_root / 'api' / 'src' / 'models'
+    print(f"Downloading models to {models_dir}")
+    models_dir.mkdir(exist_ok=True)
+    
+    # Default PTH model if no arguments provided
+    default_models = [
+        "https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.1.0/kokoro-v0_19.pth"
+    ]
+    
+    # Use provided models or default
+    models_to_download = custom_models if custom_models else default_models
+    
+    for model_url in models_to_download:
+        try:
+            download_file(model_url, models_dir)
+        except Exception as e:
+            print(f"Error downloading {model_url}: {e}")
+
+if __name__ == "__main__":
+    main(sys.argv[1:] if len(sys.argv) > 1 else None)
\ No newline at end of file
diff --git a/docker/gpu/download_pth.sh b/docker/gpu/download_pth.sh
new file mode 100755
index 0000000..c8bda83
--- /dev/null
+++ b/docker/gpu/download_pth.sh
@@ -0,0 +1,31 @@
+#!/bin/bash
+
+# Ensure models directory exists
+mkdir -p api/src/models
+
+# Function to download a file
+download_file() {
+    local url="$1"
+    local filename=$(basename "$url")
+    echo "Downloading $filename..."
+    curl -L "$url" -o "api/src/models/$filename"
+}
+
+# Default PTH model if no arguments provided
+DEFAULT_MODELS=(
+    "https://github.com/remsky/Kokoro-FastAPI/releases/download/v0.1.0/kokoro-v0_19.pth"
+)
+
+# Use provided models or default
+if [ $# -gt 0 ]; then
+    MODELS=("$@")
+else
+    MODELS=("${DEFAULT_MODELS[@]}")
+fi
+
+# Download all models
+for model in "${MODELS[@]}"; do
+    download_file "$model"
+done
+
+echo "PyTorch model download complete!"
\ No newline at end of file
diff --git a/uv.lock b/uv.lock
index 52e4de2..ab36884 100644
--- a/uv.lock
+++ b/uv.lock
@@ -2,9 +2,17 @@ version = 1
 requires-python = ">=3.10"
 resolution-markers = [
     "python_full_version == '3.11.*'",
+    "(python_full_version == '3.11.*' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.11.*' and sys_platform != 'darwin' and sys_platform != 'linux')",
+    "(python_full_version == '3.11.*' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.11.*' and sys_platform == 'darwin')",
     "python_full_version < '3.11'",
+    "(python_full_version < '3.11' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform != 'darwin' and sys_platform != 'linux')",
+    "(python_full_version < '3.11' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'darwin')",
     "python_full_version >= '3.13'",
+    "(python_full_version >= '3.13' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.13' and sys_platform != 'darwin' and sys_platform != 'linux')",
+    "(python_full_version >= '3.13' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.13' and sys_platform == 'darwin')",
     "python_full_version == '3.12.*'",
+    "(python_full_version == '3.12.*' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.12.*' and sys_platform != 'darwin' and sys_platform != 'linux')",
+    "(python_full_version == '3.12.*' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.12.*' and sys_platform == 'darwin')",
 ]
 conflicts = [[
     { package = "kokoro-fastapi", extra = "cpu" },
@@ -798,6 +806,7 @@ dependencies = [
     { name = "phonemizer" },
     { name = "pydantic" },
     { name = "pydantic-settings" },
+    { name = "pydub" },
     { name = "python-dotenv" },
     { name = "regex" },
     { name = "requests" },
@@ -812,7 +821,8 @@ dependencies = [
 
 [package.optional-dependencies]
 cpu = [
-    { name = "torch", version = "2.5.1+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" } },
+    { name = "torch", version = "2.5.1", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+    { name = "torch", version = "2.5.1+cpu", source = { registry = "https://download.pytorch.org/whl/cpu" }, marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
 ]
 gpu = [
     { name = "torch", version = "2.5.1+cu121", source = { registry = "https://download.pytorch.org/whl/cu121" } },
@@ -824,7 +834,6 @@ test = [
     { name = "pytest" },
     { name = "pytest-asyncio" },
     { name = "pytest-cov" },
-    { name = "ruff" },
 ]
 
 [package.metadata]
@@ -845,18 +854,18 @@ requires-dist = [
     { name = "phonemizer", specifier = "==3.3.0" },
     { name = "pydantic", specifier = "==2.10.4" },
     { name = "pydantic-settings", specifier = "==2.7.0" },
+    { name = "pydub", specifier = ">=0.25.1" },
     { name = "pytest", marker = "extra == 'test'", specifier = "==8.0.0" },
     { name = "pytest-asyncio", marker = "extra == 'test'", specifier = "==0.23.5" },
     { name = "pytest-cov", marker = "extra == 'test'", specifier = "==4.1.0" },
     { name = "python-dotenv", specifier = "==1.0.1" },
     { name = "regex", specifier = "==2024.11.6" },
     { name = "requests", specifier = "==2.32.3" },
-    { name = "ruff", marker = "extra == 'test'", specifier = ">=0.2.2" },
     { name = "scipy", specifier = "==1.14.1" },
     { name = "soundfile", specifier = "==0.13.0" },
     { name = "sqlalchemy", specifier = "==2.0.27" },
     { name = "tiktoken", specifier = "==0.8.0" },
-    { name = "torch", marker = "extra == 'cpu'", specifier = "==2.5.1+cpu", index = "https://download.pytorch.org/whl/cpu", conflict = { package = "kokoro-fastapi", extra = "cpu" } },
+    { name = "torch", marker = "extra == 'cpu'", specifier = "==2.5.1", index = "https://download.pytorch.org/whl/cpu", conflict = { package = "kokoro-fastapi", extra = "cpu" } },
     { name = "torch", marker = "extra == 'gpu'", specifier = "==2.5.1+cu121", index = "https://download.pytorch.org/whl/cu121", conflict = { package = "kokoro-fastapi", extra = "gpu" } },
     { name = "tqdm", specifier = "==4.67.1" },
     { name = "transformers", specifier = "==4.47.1" },
@@ -2334,24 +2343,52 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/68/4f/12207897848a653d03ebbf6775a29d949408ded5f99b2d87198bc5c93508/tomlkit-0.12.0-py3-none-any.whl", hash = "sha256:926f1f37a1587c7a4f6c7484dae538f1345d96d793d9adab5d3675957b1d0766", size = 37334 },
 ]
 
+[[package]]
+name = "torch"
+version = "2.5.1"
+source = { registry = "https://download.pytorch.org/whl/cpu" }
+resolution-markers = [
+    "(python_full_version == '3.11.*' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.11.*' and sys_platform == 'darwin')",
+    "(python_full_version < '3.11' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'darwin')",
+    "(python_full_version >= '3.13' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.13' and sys_platform == 'darwin')",
+    "(python_full_version == '3.12.*' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.12.*' and sys_platform == 'darwin')",
+]
+dependencies = [
+    { name = "filelock", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+    { name = "fsspec", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+    { name = "jinja2", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+    { name = "networkx", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+    { name = "setuptools", marker = "(python_full_version >= '3.12' and platform_machine == 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.12' and sys_platform == 'darwin')" },
+    { name = "sympy", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+    { name = "typing-extensions", marker = "(platform_machine == 'aarch64' and sys_platform == 'linux') or sys_platform == 'darwin'" },
+]
+wheels = [
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.5.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:269b10c34430aa8e9643dbe035dc525c4a9b1d671cd3dbc8ecbcaed280ae322d" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.5.1-cp310-none-macosx_11_0_arm64.whl", hash = "sha256:23d062bf70776a3d04dbe74db950db2a5245e1ba4f27208a87f0d743b0d06e86" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.5.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d5b3203f191bc40783c99488d2e776dcf93ac431a59491d627a1ca5b3ae20b22" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.5.1-cp311-none-macosx_11_0_arm64.whl", hash = "sha256:31f8c39660962f9ae4eeec995e3049b5492eb7360dd4f07377658ef4d728fa4c" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.5.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:36d1be99281b6f602d9639bd0af3ee0006e7aab16f6718d86f709d395b6f262c" },
+    { url = "https://download.pytorch.org/whl/cpu/torch-2.5.1-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:8c712df61101964eb11910a846514011f0b6f5920c55dbf567bff8a34163d5b1" },
+]
+
 [[package]]
 name = "torch"
 version = "2.5.1+cpu"
 source = { registry = "https://download.pytorch.org/whl/cpu" }
 resolution-markers = [
-    "python_full_version == '3.11.*'",
-    "python_full_version < '3.11'",
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.12.*'",
+    "(python_full_version == '3.11.*' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.11.*' and sys_platform != 'darwin' and sys_platform != 'linux')",
+    "(python_full_version < '3.11' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform != 'darwin' and sys_platform != 'linux')",
+    "(python_full_version >= '3.13' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.13' and sys_platform != 'darwin' and sys_platform != 'linux')",
+    "(python_full_version == '3.12.*' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version == '3.12.*' and sys_platform != 'darwin' and sys_platform != 'linux')",
 ]
 dependencies = [
-    { name = "filelock" },
-    { name = "fsspec" },
-    { name = "jinja2" },
-    { name = "networkx" },
-    { name = "setuptools", marker = "python_full_version >= '3.12'" },
-    { name = "sympy" },
-    { name = "typing-extensions" },
+    { name = "filelock", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "fsspec", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "jinja2", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "networkx", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "setuptools", marker = "(python_full_version >= '3.12' and platform_machine != 'aarch64' and sys_platform == 'linux') or (python_full_version >= '3.12' and sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "sympy", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
+    { name = "typing-extensions", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
 ]
 wheels = [
     { url = "https://download.pytorch.org/whl/cpu/torch-2.5.1%2Bcpu-cp310-cp310-linux_x86_64.whl", hash = "sha256:7f91a2200e352745d70e22396bd501448e28350fbdbd8d8b1c83037e25451150" },