FastDeploy/fastdeploy/entrypoints/cli/openai.py

"""
# Copyright (c) 2025  PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License"
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""

# This file is modified from https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/cli/openai.py

from __future__ import annotations

import argparse
import os
import signal
import sys
from typing import TYPE_CHECKING

from openai import OpenAI
from openai.types.chat import ChatCompletionMessageParam

from fastdeploy.entrypoints.cli.types import CLISubcommand

if TYPE_CHECKING:
    from fastdeploy.utils import FlexibleArgumentParser


def _register_signal_handlers():

    def signal_handler(sig, frame):
        sys.exit(0)

    signal.signal(signal.SIGINT, signal_handler)
    signal.signal(signal.SIGTSTP, signal_handler)


def _interactive_cli(args: argparse.Namespace) -> tuple[str, OpenAI]:
    _register_signal_handlers()

    base_url = args.url
    api_key = args.api_key or os.environ.get("OPENAI_API_KEY", "EMPTY")
    openai_client = OpenAI(api_key=api_key, base_url=base_url)

    if args.model_name:
        model_name = args.model_name
    else:
        available_models = openai_client.models.list()
        model_name = available_models.data[0].id

    print(f"Using model: {model_name}")

    return model_name, openai_client


def chat(system_prompt: str | None, model_name: str, client: OpenAI) -> None:
    conversation: list[ChatCompletionMessageParam] = []
    if system_prompt is not None:
        conversation.append({"role": "system", "content": system_prompt})

    print("Please enter a message for the chat model:")
    while True:
        try:
            input_message = input("> ")
        except EOFError:
            break
        conversation.append({"role": "user", "content": input_message})

        chat_completion = client.chat.completions.create(model=model_name, messages=conversation)

        response_message = chat_completion.choices[0].message
        output = response_message.content

        conversation.append(response_message)  # type: ignore
        print(output)


def _add_query_options(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
    parser.add_argument(
        "--url",
        type=str,
        default="http://localhost:9904/v1",
        help="url of the running OpenAI-Compatible RESTful API server",
    )
    parser.add_argument(
        "--model-name",
        type=str,
        default=None,
        help=("The model name used in prompt completion, default to " "the first model in list models API call."),
    )
    parser.add_argument(
        "--api-key",
        type=str,
        default=None,
        help=(
            "API key for OpenAI services. If provided, this api key "
            "will overwrite the api key obtained through environment variables."
        ),
    )
    return parser


class ChatCommand(CLISubcommand):
    """The `chat` subcommand for the fastdeploy CLI."""

    name = "chat"

    @staticmethod
    def cmd(args: argparse.Namespace) -> None:
        model_name, client = _interactive_cli(args)
        system_prompt = args.system_prompt
        conversation: list[ChatCompletionMessageParam] = []

        if system_prompt is not None:
            conversation.append({"role": "system", "content": system_prompt})

        if args.quick:
            conversation.append({"role": "user", "content": args.quick})

            chat_completion = client.chat.completions.create(model=model_name, messages=conversation)
            print(chat_completion.choices[0].message.content)
            return

        print("Please enter a message for the chat model:")
        while True:
            try:
                input_message = input("> ")
            except EOFError:
                break
            conversation.append({"role": "user", "content": input_message})

            chat_completion = client.chat.completions.create(model=model_name, messages=conversation)

            response_message = chat_completion.choices[0].message
            output = response_message.content

            conversation.append(response_message)  # type: ignore
            print(output)

    @staticmethod
    def add_cli_args(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
        """Add CLI arguments for the chat command."""
        _add_query_options(parser)
        parser.add_argument(
            "--system-prompt",
            type=str,
            default=None,
            help=(
                "The system prompt to be added to the chat template, " "used for models that support system prompts."
            ),
        )
        parser.add_argument(
            "-q",
            "--quick",
            type=str,
            metavar="MESSAGE",
            help=("Send a single prompt as MESSAGE " "and print the response, then exit."),
        )
        return parser

    def subparser_init(self, subparsers: argparse._SubParsersAction) -> FlexibleArgumentParser:
        parser = subparsers.add_parser(
            "chat",
            help="Generate chat completions via the running API server.",
            description="Generate chat completions via the running API server.",
            usage="fastdeploy chat [options]",
        )
        return ChatCommand.add_cli_args(parser)


class CompleteCommand(CLISubcommand):
    """The `complete` subcommand for the fastdeloy CLI."""

    name = "complete"

    @staticmethod
    def cmd(args: argparse.Namespace) -> None:
        model_name, client = _interactive_cli(args)

        if args.quick:
            completion = client.completions.create(model=model_name, prompt=args.quick)
            print(completion.choices[0].text)
            return

        print("Please enter prompt to complete:")
        while True:
            try:
                input_prompt = input("> ")
            except EOFError:
                break
            completion = client.completions.create(model=model_name, prompt=input_prompt)
            output = completion.choices[0].text
            print(output)

    @staticmethod
    def add_cli_args(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
        """Add CLI arguments for the complete command."""
        _add_query_options(parser)
        parser.add_argument(
            "-q",
            "--quick",
            type=str,
            metavar="PROMPT",
            help="Send a single prompt and print the completion output, then exit.",
        )
        return parser

    def subparser_init(self, subparsers: argparse._SubParsersAction) -> FlexibleArgumentParser:
        parser = subparsers.add_parser(
            "complete",
            help=("Generate text completions based on the given prompt " "via the running API server."),
            description=("Generate text completions based on the given prompt " "via the running API server."),
            usage="fastdeploy complete [options]",
        )
        return CompleteCommand.add_cli_args(parser)


def cmd_init() -> list[CLISubcommand]:
    return [ChatCommand(), CompleteCommand()]