MCPcopy Create free account
hub / github.com/algorithmicsuperintelligence/optillm / __init__

Method __init__

optillm/inference.py:1746–1753  ·  view source on GitHub ↗
(self)

Source from the content-addressed store, hash-verified

1744 """OpenAI SDK Compatible client for local inference with dynamic model support"""
1745
1746 def __init__(self):
1747 self.cache_manager = CacheManager.get_instance(max_size=4)
1748 self.device_manager = DeviceManager()
1749 self.model_manager = ModelManager(self.cache_manager, self.device_manager)
1750 self.lora_manager = LoRAManager(self.cache_manager)
1751 self.mlx_manager = MLXManager(self.cache_manager)
1752 self.chat = self.Chat(self)
1753 self.models = self.Models()
1754
1755 def get_pipeline(self, model: str):
1756 """Get inference pipeline - automatically chooses MLX or PyTorch based on model"""

Callers

nothing calls this directly

Calls 5

DeviceManagerClass · 0.85
ModelManagerClass · 0.85
LoRAManagerClass · 0.85
MLXManagerClass · 0.85
get_instanceMethod · 0.80

Tested by

no test coverage detected