from __future__ import annotations import logging import json import uvicorn import secrets from fastapi import FastAPI, Response, Request from fastapi.responses import StreamingResponse, RedirectResponse, HTMLResponse, JSONResponse from fastapi.exceptions import RequestValidationError from fastapi.security import APIKeyHeader from starlette.exceptions import HTTPException from starlette.status import HTTP_422_UNPROCESSABLE_ENTITY, HTTP_401_UNAUTHORIZED, HTTP_403_FORBIDDEN from fastapi.encoders import jsonable_encoder from pydantic import BaseModel from typing import Union, Optional import g4f import g4f.debug from g4f.client import AsyncClient from g4f.typing import Messages def create_app(g4f_api_key:str = None): app = FastAPI() api = Api(app, g4f_api_key=g4f_api_key) api.register_routes() api.register_authorization() api.register_validation_exception_handler() return app class ChatCompletionsConfig(BaseModel): messages: Messages model: str provider: Optional[str] = None stream: bool = False temperature: Optional[float] = None max_tokens: Optional[int] = None stop: Union[list[str], str, None] = None api_key: Optional[str] = None web_search: Optional[bool] = None proxy: Optional[str] = None list_ignored_providers: list[str] = None def set_list_ignored_providers(ignored: list[str]): global list_ignored_providers list_ignored_providers = ignored class Api: def __init__(self, app: FastAPI, g4f_api_key=None) -> None: self.app = app self.client = AsyncClient() self.g4f_api_key = g4f_api_key self.get_g4f_api_key = APIKeyHeader(name="g4f-api-key") def register_authorization(self): @self.app.middleware("http") async def authorization(request: Request, call_next): if self.g4f_api_key and request.url.path in ["/v1/chat/completions", "/v1/completions"]: try: user_g4f_api_key = await self.get_g4f_api_key(request) except HTTPException as e: if e.status_code == 403: return JSONResponse( status_code=HTTP_401_UNAUTHORIZED, content=jsonable_encoder({"detail": "G4F API key required"}), ) if not secrets.compare_digest(self.g4f_api_key, user_g4f_api_key): return JSONResponse( status_code=HTTP_403_FORBIDDEN, content=jsonable_encoder({"detail": "Invalid G4F API key"}), ) response = await call_next(request) return response def register_validation_exception_handler(self): @self.app.exception_handler(RequestValidationError) async def validation_exception_handler(request: Request, exc: RequestValidationError): details = exc.errors() modified_details = [] for error in details: modified_details.append({ "loc": error["loc"], "message": error["msg"], "type": error["type"], }) return JSONResponse( status_code=HTTP_422_UNPROCESSABLE_ENTITY, content=jsonable_encoder({"detail": modified_details}), ) def register_routes(self): @self.app.get("/") async def read_root(): return RedirectResponse("/v1", 302) @self.app.get("/v1") async def read_root_v1(): return HTMLResponse('g4f API: Go to ' 'chat/completions ' 'or models.') @self.app.get("/v1/models") async def models(): model_list = dict( (model, g4f.models.ModelUtils.convert[model]) for model in g4f.Model.__all__() ) model_list = [{ 'id': model_id, 'object': 'model', 'created': 0, 'owned_by': model.base_provider } for model_id, model in model_list.items()] return JSONResponse(model_list) @self.app.get("/v1/models/{model_name}") async def model_info(model_name: str): try: model_info = g4f.models.ModelUtils.convert[model_name] return JSONResponse({ 'id': model_name, 'object': 'model', 'created': 0, 'owned_by': model_info.base_provider }) except: return JSONResponse({"error": "The model does not exist."}) @self.app.post("/v1/chat/completions") async def chat_completions(config: ChatCompletionsConfig, request: Request = None, provider: str = None): try: config.provider = provider if config.provider is None else config.provider if config.api_key is None and request is not None: auth_header = request.headers.get("Authorization") if auth_header is not None: auth_header = auth_header.split(None, 1)[-1] if auth_header and auth_header != "Bearer": config.api_key = auth_header response = self.client.chat.completions.create( **config.dict(exclude_none=True), ignored=list_ignored_providers ) except Exception as e: logging.exception(e) return Response(content=format_exception(e, config), status_code=500, media_type="application/json") if not config.stream: return JSONResponse((await response).to_json()) async def streaming(): try: async for chunk in response: yield f"data: {json.dumps(chunk.to_json())}\n\n" except GeneratorExit: pass except Exception as e: logging.exception(e) yield f'data: {format_exception(e, config)}\n\n' yield "data: [DONE]\n\n" return StreamingResponse(streaming(), media_type="text/event-stream") @self.app.post("/v1/completions") async def completions(): return Response(content=json.dumps({'info': 'Not working yet.'}, indent=4), media_type="application/json") def format_exception(e: Exception, config: ChatCompletionsConfig) -> str: last_provider = g4f.get_last_provider(True) return json.dumps({ "error": {"message": f"{e.__class__.__name__}: {e}"}, "model": last_provider.get("model") if last_provider else config.model, "provider": last_provider.get("name") if last_provider else config.provider }) def run_api( host: str = '0.0.0.0', port: int = 1337, bind: str = None, debug: bool = False, workers: int = None, use_colors: bool = None, g4f_api_key: str = None ) -> None: print(f'Starting server... [g4f v-{g4f.version.utils.current_version}]' + (" (debug)" if debug else "")) if use_colors is None: use_colors = debug if bind is not None: host, port = bind.split(":") if debug: g4f.debug.logging = True uvicorn.run(create_app(g4f_api_key), host=host, port=int(port), workers=workers, use_colors=use_colors)