Skip to content

[V1][Structured Output] Minor modification to _validate_structured_output() #16748

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 18, 2025
+10 −10
Merged

[V1][Structured Output] Minor modification to _validate_structured_output() #16748

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 10 additions & 10 deletions vllm/v1/engine/processor.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,6 +149,7 @@ def _validate_structured_output(self, params: SamplingParams) -> None:
"xgrammar", "xgrammar:disable-any-whitespace", "guidance",
"guidance:disable-any-whitespace", "auto"
]

engine_level_backend = self.decoding_config.guided_decoding_backend
if engine_level_backend not in supported_backends:
raise ValueError(f"Only {supported_backends} structured output is "
Expand All @@ -169,8 +170,15 @@ def _validate_structured_output(self, params: SamplingParams) -> None:
if engine_level_backend.startswith("xgrammar"):
# xgrammar with no fallback
validate_xgrammar_grammar(params)
params.guided_decoding.backend = engine_level_backend
elif engine_level_backend == "auto":
elif engine_level_backend.startswith("guidance"):
# TODO: ideally we would have the LLTokenizer here as Lark syntax
# allows <|special_token|> and similar, see
# https://github.com/guidance-ai/llguidance/blob/main/docs/syntax.md#special-tokens
# Without tokenizer these are disallowed in grammars.
validate_guidance_grammar(params, tokenizer=None)
else:
# NOTE: engine_level_backend must be "auto" here, because we have
# checked supported_backends above.
# "auto" is an opt-in to opinionated behavior where we try to
# choose a backend based on request contents. This is not the
# default as it is less predictable and subject to change
Expand All @@ -183,14 +191,6 @@ def _validate_structured_output(self, params: SamplingParams) -> None:
# are not supported in xgrammar. Fall back to guidance.
params.guided_decoding.backend = "guidance"

if engine_level_backend.startswith("guidance"):
# TODO ideally we would have the LLTokenizer here as Lark syntax
# allows <|special_token|> and similar, see
# https://github.com/guidance-ai/llguidance/blob/main/docs/syntax.md#special-tokens
# Without tokenizer these are disallowed in grammars.
validate_guidance_grammar(params, tokenizer=None)
params.guided_decoding.backend = engine_level_backend

def process_inputs(
self,
request_id: str,
Expand Down