[Frontend] Gracefully handle missing chat template and fix CI failure (#7238)

Co-authored-by: Roger Wang <ywang@roblox.com>

tests/async_engine/test_chat_template.py

@@ -1,22 +1,16 @@
-import os
-import pathlib
-
 import pytest
 
-from vllm.entrypoints.chat_utils import load_chat_template
+from vllm.entrypoints.chat_utils import apply_chat_template, load_chat_template
 from vllm.entrypoints.openai.protocol import ChatCompletionRequest
 from vllm.transformers_utils.tokenizer import get_tokenizer
 
-chatml_jinja_path = pathlib.Path(os.path.dirname(os.path.abspath(
+from ..utils import VLLM_PATH
-    __file__))).parent.parent / "examples/template_chatml.jinja"
+
+chatml_jinja_path = VLLM_PATH / "examples/template_chatml.jinja"
 assert chatml_jinja_path.exists()
 
 # Define models, templates, and their corresponding expected outputs
 MODEL_TEMPLATE_GENERATON_OUTPUT = [
-    ("facebook/opt-125m", None, True,
-     "Hello</s>Hi there!</s>What is the capital of</s>"),
-    ("facebook/opt-125m", None, False,
-     "Hello</s>Hi there!</s>What is the capital of</s>"),
     ("facebook/opt-125m", chatml_jinja_path, True, """<|im_start|>user
 Hello<|im_end|>
 <|im_start|>assistant
@@ -93,11 +87,12 @@ def test_get_gen_prompt(model, template, add_generation_prompt,
         add_generation_prompt=add_generation_prompt)
 
     # Call the function and get the result
-    result = tokenizer.apply_chat_template(
+    result = apply_chat_template(
+        tokenizer,
         conversation=mock_request.messages,
-        tokenize=False,
+        chat_template=mock_request.chat_template or template_content,
         add_generation_prompt=mock_request.add_generation_prompt,
-        chat_template=mock_request.chat_template or template_content)
+    )
 
     # Test assertion
     assert result == expected_output, (

tests/async_engine/test_openapi_server_ray.py

@@ -1,10 +1,12 @@
 import openai  # use the official client for correctness check
 import pytest
 
-from ..utils import RemoteOpenAIServer
+from ..utils import VLLM_PATH, RemoteOpenAIServer
 
 # any model with a chat template should work here
 MODEL_NAME = "facebook/opt-125m"
+chatml_jinja_path = VLLM_PATH / "examples/template_chatml.jinja"
+assert chatml_jinja_path.exists()
 
 
 @pytest.fixture(scope="module")
@@ -16,7 +18,9 @@ def server():
         "--max-model-len",
         "2048",
         "--enforce-eager",
-        "--engine-use-ray"
+        "--engine-use-ray",
+        "--chat-template",
+        str(chatml_jinja_path),
     ]
 
     with RemoteOpenAIServer(MODEL_NAME, args) as remote_server:
@@ -83,7 +87,7 @@ async def test_single_chat_session(client: openai.AsyncOpenAI):
     choice = chat_completion.choices[0]
     assert choice.finish_reason == "length"
     assert chat_completion.usage == openai.types.CompletionUsage(
-        completion_tokens=10, prompt_tokens=13, total_tokens=23)
+        completion_tokens=10, prompt_tokens=55, total_tokens=65)
 
     message = choice.message
     assert message.content is not None and len(message.content) >= 10

tests/entrypoints/openai/test_oot_registration.py

@@ -9,6 +9,11 @@ from vllm.model_executor.models.opt import OPTForCausalLM
 from vllm.model_executor.sampling_metadata import SamplingMetadata
 from vllm.utils import get_open_port
 
+from ...utils import VLLM_PATH, RemoteOpenAIServer
+
+chatml_jinja_path = VLLM_PATH / "examples/template_chatml.jinja"
+assert chatml_jinja_path.exists()
+
 
 class MyOPTForCausalLM(OPTForCausalLM):
 
@@ -21,12 +26,25 @@ class MyOPTForCausalLM(OPTForCausalLM):
         return logits
 
 
-def server_function(port):
+def server_function(port: int):
     # register our dummy model
     ModelRegistry.register_model("OPTForCausalLM", MyOPTForCausalLM)
-    sys.argv = ["placeholder.py"] + \
+
-        ("--model facebook/opt-125m --gpu-memory-utilization 0.10 "
+    sys.argv = ["placeholder.py"] + [
-        f"--dtype float32 --api-key token-abc123 --port {port}").split()
+        "--model",
+        "facebook/opt-125m",
+        "--gpu-memory-utilization",
+        "0.10",
+        "--dtype",
+        "float32",
+        "--api-key",
+        "token-abc123",
+        "--port",
+        str(port),
+        "--chat-template",
+        str(chatml_jinja_path),
+    ]
+
     import runpy
     runpy.run_module('vllm.entrypoints.openai.api_server', run_name='__main__')
 
@@ -36,35 +54,40 @@ def test_oot_registration_for_api_server():
     ctx = torch.multiprocessing.get_context()
     server = ctx.Process(target=server_function, args=(port, ))
     server.start()
-    MAX_SERVER_START_WAIT_S = 60
+
-    client = OpenAI(
+    try:
-        base_url=f"http://localhost:{port}/v1",
+        client = OpenAI(
-        api_key="token-abc123",
+            base_url=f"http://localhost:{port}/v1",
-    )
+            api_key="token-abc123",
-    now = time.time()
+        )
-    while True:
+        now = time.time()
-        try:
+        while True:
-            completion = client.chat.completions.create(
+            try:
-                model="facebook/opt-125m",
+                completion = client.chat.completions.create(
-                messages=[{
+                    model="facebook/opt-125m",
-                    "role": "system",
+                    messages=[{
-                    "content": "You are a helpful assistant."
+                        "role": "system",
-                }, {
+                        "content": "You are a helpful assistant."
-                    "role": "user",
+                    }, {
-                    "content": "Hello!"
+                        "role": "user",
-                }],
+                        "content": "Hello!"
-                temperature=0,
+                    }],
-            )
+                    temperature=0,
-            break
+                )
-        except OpenAIError as e:
+                break
-            if "Connection error" in str(e):
+            except OpenAIError as e:
-                time.sleep(3)
+                if "Connection error" in str(e):
-                if time.time() - now > MAX_SERVER_START_WAIT_S:
+                    time.sleep(3)
-                    raise RuntimeError("Server did not start in time") from e
+                    if time.time() - now > RemoteOpenAIServer.MAX_START_WAIT_S:
-            else:
+                        msg = "Server did not start in time"
-                raise e
+                        raise RuntimeError(msg) from e
-    server.kill()
+                else:
+                    raise e
+    finally:
+        server.terminate()
+
     generated_text = completion.choices[0].message.content
+    assert generated_text is not None
     # make sure only the first token is generated
     rest = generated_text.replace("<s>", "")
     assert rest == ""

tests/utils.py

@@ -50,7 +50,7 @@ VLLM_PATH = Path(__file__).parent.parent
 
 class RemoteOpenAIServer:
     DUMMY_API_KEY = "token-abc123"  # vLLM's OpenAI server does not need API key
-    MAX_SERVER_START_WAIT_S = 120  # wait for server to start for 120 seconds
+    MAX_START_WAIT_S = 120  # wait for server to start for 120 seconds
 
     def __init__(
         self,
@@ -85,7 +85,7 @@ class RemoteOpenAIServer:
                                      stdout=sys.stdout,
                                      stderr=sys.stderr)
         self._wait_for_server(url=self.url_for("health"),
-                              timeout=self.MAX_SERVER_START_WAIT_S)
+                              timeout=self.MAX_START_WAIT_S)
 
     def __enter__(self):
         return self

vllm/entrypoints/chat_utils.py

@@ -1,8 +1,9 @@
 import codecs
 from dataclasses import dataclass
 from functools import lru_cache
-from typing import (Awaitable, Iterable, List, Optional, Tuple, Union, cast,
+from pathlib import Path
-                    final)
+from typing import (Any, Awaitable, Iterable, List, Optional, Tuple, Union,
+                    cast, final)
 
 # yapf conflicts with isort for this block
 # yapf: disable
@@ -22,6 +23,7 @@ from vllm.config import ModelConfig
 from vllm.logger import init_logger
 from vllm.multimodal import MultiModalDataDict
 from vllm.multimodal.utils import async_get_and_parse_image
+from vllm.transformers_utils.tokenizer import AnyTokenizer
 
 logger = init_logger(__name__)
 
@@ -69,13 +71,17 @@ class ChatMessageParseResult:
     mm_futures: List[Awaitable[MultiModalDataDict]]
 
 
-def load_chat_template(chat_template: Optional[str]) -> Optional[str]:
+def load_chat_template(
+        chat_template: Optional[Union[Path, str]]) -> Optional[str]:
     if chat_template is None:
         return None
     try:
         with open(chat_template, "r") as f:
             resolved_chat_template = f.read()
     except OSError as e:
+        if isinstance(chat_template, Path):
+            raise
+
         JINJA_CHARS = "{}\n"
         if not any(c in chat_template for c in JINJA_CHARS):
             msg = (f"The supplied chat template ({chat_template}) "
@@ -208,3 +214,28 @@ def parse_chat_messages(
         mm_futures.extend(parse_result.mm_futures)
 
     return conversation, mm_futures
+
+
+def apply_chat_template(
+    tokenizer: AnyTokenizer,
+    conversation: List[ConversationMessage],
+    chat_template: Optional[str],
+    *,
+    tokenize: bool = False,  # Different from HF's default
+    **kwargs: Any,
+) -> str:
+    if chat_template is None and tokenizer.chat_template is None:
+        raise ValueError(
+            "As of transformers v4.44, default chat template is no longer "
+            "allowed, so you must provide a chat template if the tokenizer "
+            "does not define one.")
+
+    prompt = tokenizer.apply_chat_template(
+        conversation=conversation,
+        chat_template=chat_template,
+        tokenize=tokenize,
+        **kwargs,
+    )
+    assert isinstance(prompt, str)
+
+    return prompt

vllm/entrypoints/openai/protocol.py

@@ -190,8 +190,9 @@ class ChatCompletionRequest(OpenAIBaseModel):
         default=None,
         description=(
             "A Jinja template to use for this conversion. "
-            "If this is not passed, the model's default chat template will be "
+            "As of transformers v4.44, default chat template is no longer "
-            "used instead."),
+            "allowed, so you must provide a chat template if the tokenizer "
+            "does not define one."),
     )
     chat_template_kwargs: Optional[Dict[str, Any]] = Field(
         default=None,

vllm/entrypoints/openai/serving_chat.py

@@ -10,6 +10,7 @@ from transformers import PreTrainedTokenizer
 from vllm.config import ModelConfig
 from vllm.engine.protocol import AsyncEngineClient
 from vllm.entrypoints.chat_utils import (ConversationMessage,
+                                         apply_chat_template,
                                          load_chat_template,
                                          parse_chat_messages)
 from vllm.entrypoints.logger import RequestLogger
@@ -99,16 +100,15 @@ class OpenAIServingChat(OpenAIServing):
                 tool.model_dump() for tool in request.tools
             ]
 
-            prompt = tokenizer.apply_chat_template(
+            prompt = apply_chat_template(
+                tokenizer,
                 conversation=conversation,
-                tokenize=False,
+                chat_template=request.chat_template or self.chat_template,
                 add_generation_prompt=request.add_generation_prompt,
                 tools=tool_dicts,
                 documents=request.documents,
-                chat_template=request.chat_template or self.chat_template,
                 **(request.chat_template_kwargs or {}),
             )
-            assert isinstance(prompt, str)
         except Exception as e:
             logger.error("Error in applying chat template from request: %s", e)
             return self.create_error_response(str(e))

vllm/entrypoints/openai/serving_tokenization.py

@@ -2,7 +2,9 @@ from typing import List, Optional, Union
 
 from vllm.config import ModelConfig
 from vllm.engine.protocol import AsyncEngineClient
-from vllm.entrypoints.chat_utils import load_chat_template, parse_chat_messages
+from vllm.entrypoints.chat_utils import (apply_chat_template,
+                                         load_chat_template,
+                                         parse_chat_messages)
 from vllm.entrypoints.logger import RequestLogger
 # yapf conflicts with isort for this block
 # yapf: disable
@@ -70,12 +72,12 @@ class OpenAIServingTokenization(OpenAIServing):
                 logger.warning(
                     "Multi-modal inputs are ignored during tokenization")
 
-            prompt = tokenizer.apply_chat_template(
+            prompt = apply_chat_template(
-                add_generation_prompt=request.add_generation_prompt,
+                tokenizer,
                 conversation=conversation,
-                tokenize=False,
+                chat_template=self.chat_template,
-                chat_template=self.chat_template)
+                add_generation_prompt=request.add_generation_prompt,
-            assert isinstance(prompt, str)
+            )
         else:
             prompt = request.prompt
 

vllm/transformers_utils/tokenizer.py

@@ -12,12 +12,12 @@ from vllm.lora.request import LoRARequest
 from vllm.transformers_utils.tokenizers import BaichuanTokenizer
 from vllm.utils import make_async
 
+from .tokenizer_group import AnyTokenizer
+
 logger = init_logger(__name__)
 
 
-def get_cached_tokenizer(
+def get_cached_tokenizer(tokenizer: AnyTokenizer) -> AnyTokenizer:
-    tokenizer: Union[PreTrainedTokenizer, PreTrainedTokenizerFast]
-) -> Union[PreTrainedTokenizer, PreTrainedTokenizerFast]:
     """Get tokenizer with cached properties.
 
     This will patch the tokenizer object in place.
@@ -63,7 +63,7 @@ def get_tokenizer(
     revision: Optional[str] = None,
     download_dir: Optional[str] = None,
     **kwargs,
-) -> Union[PreTrainedTokenizer, PreTrainedTokenizerFast]:
+) -> AnyTokenizer:
     """Gets a tokenizer for the given model name via HuggingFace or ModelScope.
     """
     if VLLM_USE_MODELSCOPE: