Method load_primary

core/loader_v2.py:309–341

Load the primary (7B) model.

(self)

        self._load_failures: int = 0

    def load_primary(self) -> bool:
        """Load the primary (7B) model."""
        try:
            info(f"Loading model: {MODEL_PATH.name}")

            # Check if model file exists
            if not MODEL_PATH.exists():
                error(f"Model file not found: {MODEL_PATH}")
                self._load_failures += 1
                return False

            # Check if llama-server binary exists
            llama_bin = Path(LLAMA_SERVER_BIN)
            if not llama_bin.exists():
                error(f"llama-server not found: {LLAMA_SERVER_BIN}")
                self._load_failures += 1
                return False

            # Start server
            self._server = LlamaServer(MODEL_PATH)
            if not self._server.start():
                self._load_failures += 1
                return False

            self._loaded = True
            self._loaded_at = time.time()
            success(f"Loaded model ({MODEL_PATH.name})")
            return True

        except Exception as e:
            error(f"Failed to load model: {e}")
            self._load_failures += 1
            return False

    def unload(self):
        """Unload (stop) the current model server."""