File size: 11,253 Bytes
690332d
75da468
 
 
690332d
75da468
 
 
690332d
 
5491dc5
690332d
 
75da468
 
 
690332d
75da468
690332d
 
5491dc5
75da468
8f99309
 
75da468
 
 
8f99309
 
 
 
 
 
75da468
 
 
5491dc5
 
690332d
75da468
5491dc5
8f99309
690332d
 
 
5491dc5
 
75da468
 
 
5491dc5
75da468
 
 
 
690332d
 
 
 
 
 
 
 
 
 
 
75da468
 
 
690332d
 
 
 
5491dc5
690332d
8f99309
5491dc5
 
 
 
 
 
690332d
 
5491dc5
690332d
 
75da468
690332d
 
 
5491dc5
8f99309
5491dc5
8f99309
 
 
 
 
 
 
 
5491dc5
8f99309
5491dc5
8f99309
 
 
 
 
 
 
 
cf74192
690332d
5491dc5
690332d
 
 
5491dc5
75da468
8f99309
75da468
 
 
cf74192
690332d
 
 
 
 
 
 
 
 
 
 
75da468
690332d
 
 
cf74192
690332d
 
 
 
 
 
 
 
 
 
 
 
 
cf74192
75da468
 
 
 
 
 
 
 
 
 
5491dc5
 
 
 
 
 
690332d
 
 
 
 
 
 
 
 
5491dc5
 
 
 
 
 
75da468
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5491dc5
 
 
 
 
 
8f99309
75da468
690332d
 
75da468
690332d
75da468
690332d
5491dc5
8f99309
5491dc5
 
 
 
75da468
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5491dc5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
75da468
5491dc5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
690332d
5491dc5
 
 
 
 
 
 
 
690332d
 
 
75da468
690332d
 
75da468
 
5491dc5
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
import asyncio
import importlib
import inspect
import re
from contextlib import asynccontextmanager
from http import HTTPStatus
from typing import Optional, Set

import fastapi
import uvicorn
from fastapi import APIRouter, Request
from fastapi.exceptions import RequestValidationError
from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import JSONResponse, Response, StreamingResponse
from prometheus_client import make_asgi_app
from starlette.routing import Mount

import vllm.envs as envs
from vllm.engine.arg_utils import AsyncEngineArgs
from vllm.engine.async_llm_engine import AsyncLLMEngine
from vllm.entrypoints.logger import RequestLogger
from vllm.entrypoints.openai.cli_args import make_arg_parser
# yapf conflicts with isort for this block
# yapf: disable
from vllm.entrypoints.openai.protocol import (ChatCompletionRequest,
                                              ChatCompletionResponse,
                                              CompletionRequest,
                                              DetokenizeRequest,
                                              DetokenizeResponse,
                                              EmbeddingRequest, ErrorResponse,
                                              TokenizeRequest,
                                              TokenizeResponse)
# yapf: enable
from vllm.entrypoints.openai.serving_chat import OpenAIServingChat
from vllm.entrypoints.openai.serving_completion import OpenAIServingCompletion
from vllm.entrypoints.openai.serving_embedding import OpenAIServingEmbedding
from vllm.entrypoints.openai.serving_tokenization import (
    OpenAIServingTokenization)
from vllm.logger import init_logger
from vllm.usage.usage_lib import UsageContext
from vllm.utils import FlexibleArgumentParser
from vllm.version import __version__ as VLLM_VERSION

TIMEOUT_KEEP_ALIVE = 5  # seconds

engine: AsyncLLMEngine
engine_args: AsyncEngineArgs
openai_serving_chat: OpenAIServingChat
openai_serving_completion: OpenAIServingCompletion
openai_serving_embedding: OpenAIServingEmbedding
openai_serving_tokenization: OpenAIServingTokenization

logger = init_logger('vllm.entrypoints.openai.api_server')

_running_tasks: Set[asyncio.Task] = set()


@asynccontextmanager
async def lifespan(app: fastapi.FastAPI):

    async def _force_log():
        while True:
            await asyncio.sleep(10)
            await engine.do_log_stats()

    if not engine_args.disable_log_stats:
        task = asyncio.create_task(_force_log())
        _running_tasks.add(task)
        task.add_done_callback(_running_tasks.remove)

    yield


router = APIRouter()


def mount_metrics(app: fastapi.FastAPI):
    # Add prometheus asgi middleware to route /metrics requests
    metrics_route = Mount("/metrics", make_asgi_app())
    # Workaround for 307 Redirect for /metrics
    metrics_route.path_regex = re.compile('^/metrics(?P<path>.*)$')
    app.routes.append(metrics_route)


@router.get("/health")
async def health() -> Response:
    """Health check."""
    await openai_serving_chat.engine.check_health()
    return Response(status_code=200)


@router.post("/tokenize")
async def tokenize(request: TokenizeRequest):
    generator = await openai_serving_tokenization.create_tokenize(request)
    if isinstance(generator, ErrorResponse):
        return JSONResponse(content=generator.model_dump(),
                            status_code=generator.code)
    else:
        assert isinstance(generator, TokenizeResponse)
        return JSONResponse(content=generator.model_dump())


@router.post("/detokenize")
async def detokenize(request: DetokenizeRequest):
    generator = await openai_serving_tokenization.create_detokenize(request)
    if isinstance(generator, ErrorResponse):
        return JSONResponse(content=generator.model_dump(),
                            status_code=generator.code)
    else:
        assert isinstance(generator, DetokenizeResponse)
        return JSONResponse(content=generator.model_dump())


@router.get("/api/v1/models")
async def show_available_models():
    models = await openai_serving_completion.show_available_models()
    return JSONResponse(content=models.model_dump())


@router.get("/version")
async def show_version():
    ver = {"version": VLLM_VERSION}
    return JSONResponse(content=ver)


@router.post("/api/v1/chat/completions")
async def create_chat_completion(request: ChatCompletionRequest,
                                 raw_request: Request):
    generator = await openai_serving_chat.create_chat_completion(
        request, raw_request)
    if isinstance(generator, ErrorResponse):
        return JSONResponse(content=generator.model_dump(),
                            status_code=generator.code)
    if request.stream:
        return StreamingResponse(content=generator,
                                 media_type="text/event-stream")
    else:
        assert isinstance(generator, ChatCompletionResponse)
        return JSONResponse(content=generator.model_dump())


@router.post("/api/v1/completions")
async def create_completion(request: CompletionRequest, raw_request: Request):
    generator = await openai_serving_completion.create_completion(
        request, raw_request)
    if isinstance(generator, ErrorResponse):
        return JSONResponse(content=generator.model_dump(),
                            status_code=generator.code)
    if request.stream:
        return StreamingResponse(content=generator,
                                 media_type="text/event-stream")
    else:
        return JSONResponse(content=generator.model_dump())


@router.post("/api/v1/embeddings")
async def create_embedding(request: EmbeddingRequest, raw_request: Request):
    generator = await openai_serving_embedding.create_embedding(
        request, raw_request)
    if isinstance(generator, ErrorResponse):
        return JSONResponse(content=generator.model_dump(),
                            status_code=generator.code)
    else:
        return JSONResponse(content=generator.model_dump())


def build_app(args):
    app = fastapi.FastAPI(lifespan=lifespan)
    app.include_router(router)
    app.root_path = args.root_path

    mount_metrics(app)

    app.add_middleware(
        CORSMiddleware,
        allow_origins=args.allowed_origins,
        allow_credentials=args.allow_credentials,
        allow_methods=args.allowed_methods,
        allow_headers=args.allowed_headers,
    )

    @app.exception_handler(RequestValidationError)
    async def validation_exception_handler(_, exc):
        err = openai_serving_chat.create_error_response(message=str(exc))
        return JSONResponse(err.model_dump(),
                            status_code=HTTPStatus.BAD_REQUEST)

    if token := envs.VLLM_API_KEY or args.api_key:

        @app.middleware("http")
        async def authentication(request: Request, call_next):
            root_path = "" if args.root_path is None else args.root_path
            if request.method == "OPTIONS":
                return await call_next(request)
            if not request.url.path.startswith(f"{root_path}/v1"):
                return await call_next(request)
            if request.headers.get("Authorization") != "Bearer " + token:
                return JSONResponse(content={"error": "Unauthorized"},
                                    status_code=401)
            return await call_next(request)

    for middleware in args.middleware:
        module_path, object_name = middleware.rsplit(".", 1)
        imported = getattr(importlib.import_module(module_path), object_name)
        if inspect.isclass(imported):
            app.add_middleware(imported)
        elif inspect.iscoroutinefunction(imported):
            app.middleware("http")(imported)
        else:
            raise ValueError(f"Invalid middleware {middleware}. "
                             f"Must be a function or a class.")

    return app


def run_server(args, llm_engine=None):
    app = build_app(args)

    logger.info("vLLM API server version %s", VLLM_VERSION)
    logger.info("args: %s", args)

    if args.served_model_name is not None:
        served_model_names = args.served_model_name
    else:
        served_model_names = [args.model]

    global engine, engine_args

    engine_args = AsyncEngineArgs.from_cli_args(args)
    engine = (llm_engine
              if llm_engine is not None else AsyncLLMEngine.from_engine_args(
                  engine_args, usage_context=UsageContext.OPENAI_API_SERVER))

    event_loop: Optional[asyncio.AbstractEventLoop]
    try:
        event_loop = asyncio.get_running_loop()
    except RuntimeError:
        event_loop = None

    if event_loop is not None and event_loop.is_running():
        # If the current is instanced by Ray Serve,
        # there is already a running event loop
        model_config = event_loop.run_until_complete(engine.get_model_config())
    else:
        # When using single vLLM without engine_use_ray
        model_config = asyncio.run(engine.get_model_config())

    if args.disable_log_requests:
        request_logger = None
    else:
        request_logger = RequestLogger(max_log_len=args.max_log_len)

    global openai_serving_chat
    global openai_serving_completion
    global openai_serving_embedding
    global openai_serving_tokenization

    openai_serving_chat = OpenAIServingChat(
        engine,
        model_config,
        served_model_names,
        args.response_role,
        lora_modules=args.lora_modules,
        prompt_adapters=args.prompt_adapters,
        request_logger=request_logger,
        chat_template=args.chat_template,
    )
    openai_serving_completion = OpenAIServingCompletion(
        engine,
        model_config,
        served_model_names,
        lora_modules=args.lora_modules,
        prompt_adapters=args.prompt_adapters,
        request_logger=request_logger,
    )
    openai_serving_embedding = OpenAIServingEmbedding(
        engine,
        model_config,
        served_model_names,
        request_logger=request_logger,
    )
    openai_serving_tokenization = OpenAIServingTokenization(
        engine,
        model_config,
        served_model_names,
        lora_modules=args.lora_modules,
        request_logger=request_logger,
        chat_template=args.chat_template,
    )
    app.root_path = args.root_path

    logger.info("Available routes are:")
    for route in app.routes:
        if not hasattr(route, 'methods'):
            continue
        methods = ', '.join(route.methods)
        logger.info("Route: %s, Methods: %s", route.path, methods)

    uvicorn.run(app,
                host=args.host,
                port=args.port,
                log_level=args.uvicorn_log_level,
                timeout_keep_alive=TIMEOUT_KEEP_ALIVE,
                ssl_keyfile=args.ssl_keyfile,
                ssl_certfile=args.ssl_certfile,
                ssl_ca_certs=args.ssl_ca_certs,
                ssl_cert_reqs=args.ssl_cert_reqs)


if __name__ == "__main__":
    # NOTE(simon):
    # This section should be in sync with vllm/scripts.py for CLI entrypoints.
    parser = FlexibleArgumentParser(
        description="vLLM OpenAI-Compatible RESTful API server.")
    parser = make_arg_parser(parser)
    args = parser.parse_args()
    run_server(args)