xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-13 23:55:44 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Misc] Replace TODO in serving transcription (#18895)

Browse Source 
Signed-off-by: NickLucche <nlucches@redhat.com>
This commit is contained in:
Nicolò Lucchesi 2025-05-29 11:58:14 +02:00 committed by GitHub
parent 7fcfd954ff
commit 24d0ef8970
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/entrypoints/openai/serving_transcription.py
View File

@ -278,7 +278,9 @@ class OpenAIServingTranscription(OpenAIServing):
result_generator: Optional[AsyncGenerator[RequestOutput, None]] = None result_generator: Optional[AsyncGenerator[RequestOutput, None]] = None
try: try:
# TODO(rob): subtract len of tokenized prompt. # Unlike most decoder-only models, whisper generation length is not
# constrained by the size of the input audio, which is mapped to a
# fixed-size log-mel-spectogram.
default_max_tokens = self.model_config.max_model_len default_max_tokens = self.model_config.max_model_len
sampling_params = request.to_sampling_params( sampling_params = request.to_sampling_params(
default_max_tokens, self.default_sampling_params) default_max_tokens, self.default_sampling_params)