Initial commit: LocalPilot：本地模型运行时，复用 llama-server 并提供 Ollama 兼容 provider

localpilot/backends/base.py

@@ -0,0 +1,39 @@
+from __future__ import annotations
+
+from abc import ABC, abstractmethod
+from collections.abc import AsyncIterator
+from typing import Any
+
+from ..config import ModelConfig, RuntimeConfig
+from ..types import BackendError, ChatChunk, ChatParams
+
+
+class BaseBackend(ABC):
+    def __init__(self, model: ModelConfig, runtime: RuntimeConfig) -> None:
+        self.model = model
+        self.runtime = runtime
+        self.loaded = False
+
+    @abstractmethod
+    async def load(self) -> None:
+        raise NotImplementedError
+
+    @abstractmethod
+    async def unload(self) -> None:
+        raise NotImplementedError
+
+    @abstractmethod
+    async def chat(
+        self,
+        messages: list[dict[str, Any]],
+        params: ChatParams,
+        stream: bool = False,
+    ) -> AsyncIterator[ChatChunk]:
+        raise NotImplementedError
+
+    @abstractmethod
+    def info(self) -> dict[str, Any]:
+        raise NotImplementedError
+
+    async def embed(self, inputs: list[str]) -> list[list[float]]:
+        raise BackendError(f"后端 {self.model.id} 不支持 embeddings")