Support arbitrary json_object in OpenAI and Context Free Grammar (#3211)

tests/entrypoints/test_openai_server.py

@@ -660,5 +660,55 @@ async def test_guided_decoding_type_error(server, client: openai.AsyncOpenAI):
             extra_body=dict(guided_regex=TEST_REGEX, guided_json=TEST_SCHEMA))
 
 
+async def test_response_format_json_object(server, client: openai.AsyncOpenAI):
+    resp = await client.chat.completions.create(
+        model=MODEL_NAME,
+        messages=[{
+            "role":
+            "user",
+            "content": ('what is 1+1? please respond with a JSON object, '
+                        'the format is {"result": 2}')
+        }],
+        response_format={"type": "json_object"})
+
+    content = resp.choices[0].message.content
+    loaded = json.loads(content)
+    assert loaded == {"result": 2}, loaded
+
+
+async def test_guided_grammar(server, client: openai.AsyncOpenAI):
+    simple_sql_grammar = """
+start: select_statement
+
+select_statement: "SELECT" column "from" table "where" condition
+
+column: "col_1" | "col_2"
+table: "table_1" | "table_2"
+condition: column "=" number
+
+number: "1" | "2"
+"""
+
+    completion = await client.completions.create(
+        model=MODEL_NAME,
+        prompt=("Generate a sql state that select col_1 from "
+                "table_1 where it is equals to 1"),
+        temperature=1.0,
+        max_tokens=500,
+        extra_body=dict(guided_grammar=simple_sql_grammar))
+
+    content = completion.choices[0].text
+
+    # use Lark to parse the output, and make sure it's a valid parse tree
+    from lark import Lark
+    parser = Lark(simple_sql_grammar)
+    parser.parse(content)
+
+    # remove spaces for comparison b/c we removed them in the grammar
+    ground_truth = "SELECT col_1 from table_1 where col_1 = 1".replace(" ", "")
+
+    assert content.strip() == ground_truth
+
+
 if __name__ == "__main__":
     pytest.main([__file__])

vllm/entrypoints/openai/protocol.py

@@ -55,6 +55,11 @@ class UsageInfo(BaseModel):
     completion_tokens: Optional[int] = 0
 
 
+class ResponseFormat(BaseModel):
+    # type must be "json_object" or "text"
+    type: str = Literal["text", "json_object"]
+
+
 class ChatCompletionRequest(BaseModel):
     model: str
     messages: List[Dict[str, str]]
@@ -89,6 +94,8 @@ class ChatCompletionRequest(BaseModel):
     guided_json: Optional[Union[str, dict, BaseModel]] = None
     guided_regex: Optional[str] = None
     guided_choice: Optional[List[str]] = None
+    guided_grammar: Optional[str] = None
+    response_format: Optional[ResponseFormat] = None
 
     def to_sampling_params(self) -> SamplingParams:
         if self.logprobs and not self.top_logprobs:
@@ -183,6 +190,8 @@ class CompletionRequest(BaseModel):
     guided_json: Optional[Union[str, dict, BaseModel]] = None
     guided_regex: Optional[str] = None
     guided_choice: Optional[List[str]] = None
+    guided_grammar: Optional[str] = None
+    response_format: Optional[ResponseFormat] = None
 
     def to_sampling_params(self):
         echo_without_generation = self.echo and self.max_tokens == 0

vllm/model_executor/guided_decoding.py

@@ -6,20 +6,51 @@ from functools import lru_cache
 from json import dumps as json_dumps
 from re import escape as regex_escape
 from typing import Union, Tuple
+
 from pydantic import BaseModel
+from transformers import PreTrainedTokenizerBase
 
 from vllm.entrypoints.openai.protocol import (CompletionRequest,
                                               ChatCompletionRequest)
 from vllm.model_executor.guided_logits_processors import (JSONLogitsProcessor,
-                                                          RegexLogitsProcessor)
+                                                          RegexLogitsProcessor,
+                                                          CFGLogitsProcessor)
 
 
 class GuidedDecodingMode(Enum):
     JSON = "json"
     REGEX = "regex"
     CHOICE = "choice"
+    GRAMMAR = "grammar"
 
 
+# https://github.com/outlines-dev/outlines/blob/main/outlines/grammars/json.lark
+# the main difference is that we changed the start: value to
+# start: object | array, so we are denying scalar values as the root of the
+# JSON. Starting with scalars as the root seems to cause llama to generate
+# without stop.
+JSON_GRAMMAR = r"""
+?start: object | array
+
+?value: object
+| array
+| UNESCAPED_STRING
+| SIGNED_NUMBER      -> number
+| "true"             -> true
+| "false"            -> false
+| "null"             -> null
+
+array  : "[" [value ("," value)*] "]"
+object : "{" [pair ("," pair)*] "}"
+pair   : UNESCAPED_STRING ":" value
+
+%import common.UNESCAPED_STRING
+%import common.SIGNED_NUMBER
+%import common.WS
+
+%ignore WS
+"""
+
 global_thread_pool = None  # used for generating logits processor fsm
 
 
@@ -57,9 +88,6 @@ def _get_guide_and_mode(
 ) -> Tuple[str, GuidedDecodingMode]:
 
     if request.guided_json:
-        if not isinstance(request.guided_json, (str, dict, BaseModel)):
-            raise TypeError("JSON schema must be str, dict, or BaseModel")
-
         json = request.guided_json
         if isinstance(json, dict):
             # turn dict into hashable string
@@ -69,33 +97,33 @@ def _get_guide_and_mode(
             # with the same fields will get hashed the same
             json = str(json.__signature__)
         return json, GuidedDecodingMode.JSON
-
     elif request.guided_regex:
-        if not isinstance(request.guided_regex, str):
-            raise TypeError("Regex must be string")
         return request.guided_regex, GuidedDecodingMode.REGEX
-
     elif request.guided_choice:
-        if not isinstance(request.guided_choice, list):
-            raise TypeError("Choices must be a list")
-
         # choice just uses regex
         choices = [
             regex_escape(str(choice)) for choice in request.guided_choice
         ]
         choices_regex = "(" + "|".join(choices) + ")"
         return choices_regex, GuidedDecodingMode.CHOICE
-
+    elif request.guided_grammar:
+        return request.guided_grammar, GuidedDecodingMode.GRAMMAR
+    elif (request.response_format is not None
+          and request.response_format.type == "json_object"):
+        return JSON_GRAMMAR, GuidedDecodingMode.GRAMMAR
     else:
         return None, None
 
 
 @lru_cache(maxsize=32)
-def _get_cached_logits_processor(guide: str, tokenizer,
+def _get_cached_logits_processor(guide: str,
+                                 tokenizer: PreTrainedTokenizerBase,
                                  mode: GuidedDecodingMode):
     if mode == GuidedDecodingMode.JSON:
         return JSONLogitsProcessor(guide, tokenizer)
     elif mode == GuidedDecodingMode.REGEX or mode == GuidedDecodingMode.CHOICE:
         return RegexLogitsProcessor(guide, tokenizer)
+    elif mode == GuidedDecodingMode.GRAMMAR:
+        return CFGLogitsProcessor(guide, tokenizer)
     else:
         raise ValueError(f"Unknown guided decoding mode {mode}")

vllm/model_executor/guided_logits_processors.py

@@ -16,30 +16,60 @@
 import json
 import math
 from collections import defaultdict
-from typing import Union, DefaultDict, Dict, List, Optional
+from typing import Union, DefaultDict, Dict, List, Optional, Callable
 
 import torch
 from pydantic import BaseModel
-from outlines.fsm.fsm import RegexFSM
+from transformers import PreTrainedTokenizerBase
+from outlines.fsm.fsm import RegexFSM, CFGFSM
 from outlines.fsm.json_schema import build_regex_from_schema
 
 
-class RegexLogitsProcessor:
+class BaseLogitsProcessor:
 
-    def __init__(self, regex_string: str, tokenizer):
+    def adapt_tokenizer(self, tokenizer: PreTrainedTokenizerBase):
-        """Compile the FSM that drives the regex-structured generation.
+        """Adapt vLLM's tokenizer to use to compile the FSM.
 
-        Parameters
+        The API of Outlines tokenizers is slightly different to that of
-        ----------
+        `transformers`. The decoder of outlines, returns a list whereas
-        regex_string
+        the decode of vLLM returns an str. To sync the vLLM decoder with
-            A string that represents a regular expression
+        outlines internal api, the decoder should be adapted. In addition
-        tokenizer
+        we need to handle the missing spaces to Llama's tokenizer to be
-            The model's tokenizer
+        able to compile FSMs for this model.
 
         """
-        tokenizer = self.adapt_tokenizer(tokenizer)
+        if getattr(tokenizer, "_outlines_adapted", False):
-        fsm = RegexFSM(regex_string, tokenizer)
+            return tokenizer
-        self.fsm = fsm
+
+        tokenizer.vocabulary = tokenizer.get_vocab()
+        tokenizer.special_tokens = set(tokenizer.all_special_tokens)
+
+        def convert_token_to_string(token: str) -> str:
+            from transformers.file_utils import SPIECE_UNDERLINE
+
+            string = tokenizer.convert_tokens_to_string([token])
+
+            # A hack to handle missing spaces to HF's Llama tokenizers
+            if token.startswith(SPIECE_UNDERLINE) or token == "<0x20>":
+                return " " + string
+
+            return string
+
+        def change_decoder(
+            decoder: Callable[[List[int]], str]
+        ) -> Callable[[List[int]], List[str]]:
+            """Sync vLLM's decoder with the outlines by returning list."""
+
+            def new_decoder(inp_tokens: List[int]) -> List[str]:
+                return [decoder(inp_tokens)]
+
+            return new_decoder
+
+        tokenizer.convert_token_to_string = convert_token_to_string
+        tokenizer.decode = change_decoder(tokenizer.decode)
+        setattr(tokenizer, "_outlines_adapted", True)  # noqa: B010
+
+        return tokenizer
 
     def init_state(self):
         """Initialize the FSM states."""
@@ -69,38 +99,30 @@ class RegexLogitsProcessor:
 
         return scores
 
-    def adapt_tokenizer(self, tokenizer):
-        """Adapt vLLM's tokenizer to use to compile the FSM.
 
-        The API of Outlines tokenizers is slightly different to that of
+class RegexLogitsProcessor(BaseLogitsProcessor):
-        `transformers`. In addition we need to handle the missing spaces to
+
-        Llama's tokenizer to be able to compile FSMs for this model.
+    def __init__(self, regex_string: str, tokenizer: PreTrainedTokenizerBase):
+        """Compile the FSM that drives the regex-structured generation.
+
+        Parameters
+        ----------
+        regex_string
+            A string that represents a regular expression
+        tokenizer
+            The model's tokenizer
 
         """
-        tokenizer.vocabulary = tokenizer.get_vocab()
+        tokenizer = self.adapt_tokenizer(tokenizer)
-        tokenizer.special_tokens = set(tokenizer.all_special_tokens)
+        fsm = RegexFSM(regex_string, tokenizer)
-
+        self.fsm = fsm
-        def convert_token_to_string(token: str) -> str:
-            from transformers.file_utils import SPIECE_UNDERLINE
-
-            string = tokenizer.convert_tokens_to_string([token])
-
-            # A hack to handle missing spaces to HF's Llama tokenizers
-            if token.startswith(SPIECE_UNDERLINE) or token == "<0x20>":
-                return " " + string
-
-            return string
-
-        tokenizer.convert_token_to_string = convert_token_to_string
-
-        return tokenizer
 
 
 class JSONLogitsProcessor(RegexLogitsProcessor):
 
     def __init__(self,
                  schema: Union[str, Dict, BaseModel],
-                 tokenizer,
+                 tokenizer: PreTrainedTokenizerBase,
                  whitespace_pattern: Optional[str] = None):
         """Compile the FSM that drives the JSON-guided generation.
 
@@ -130,3 +152,21 @@ class JSONLogitsProcessor(RegexLogitsProcessor):
                 f"the JSON Schema specification")
         regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
         super().__init__(regex_string, tokenizer)
+
+
+class CFGLogitsProcessor(BaseLogitsProcessor):
+
+    def __init__(self, cfg: str, tokenizer: PreTrainedTokenizerBase):
+        """Compile the FSM that drives the context free grammar generation.
+
+        Parameters
+        ----------
+        cfg
+            A string that represents a context-free grammar
+        tokenizer
+            The model's tokenizer
+
+        """
+        tokenizer = self.adapt_tokenizer(tokenizer)
+        fsm = CFGFSM(cfg, tokenizer)
+        self.fsm = fsm