[Frontend] Add the Beam reasoning and tool parsers (#62052)

Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: Bugen Zhao <i@bugenzhao.com>
Co-authored-by: Yoni Gozlan <74535834+yonigozlan@users.noreply.github.com>
Co-authored-by: neilkg <33635204+neilkg@users.noreply.github.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Woosuk Kwon
2026-10-16 14:08:29 -07:00
committed by GitHub
co-authored by Bugen Zhao Yoni Gozlan neilkg Cursor Claude Opus 5.5
parent 9731f30a03
commit 153a17166e
27 changed files with 4464 additions and 92 deletions
+8 -3
View File
@@ -60,6 +60,7 @@ from vllm.renderers.chat_utils import (
has_non_text_content,
)
from vllm.renderers.inputs.preprocess import (
extract_prompt_components,
parse_model_prompt,
prompt_to_seq,
)
@@ -844,13 +845,17 @@ class OnlineRenderer:
f"but got {type(request).__name__}"
)
raise NotImplementedError(msg)
request = parser(
request_parser = parser(
tokenizer,
request.tools,
model_config=self.model_config,
chat_template_kwargs=chat_params.chat_template_kwargs,
).adjust_request(
request=request,
)
prompt_token_ids = extract_prompt_components(
self.model_config, engine_input
).token_ids
if prompt_token_ids is not None:
request_parser.set_prompt_token_ids(prompt_token_ids)
request = request_parser.adjust_request(request=request)
return conversation, [engine_input]