Simplify (and fix) passing of guided decoding backend options (#17008)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
@@ -615,9 +615,9 @@ class MQLLMEngineClient(EngineClient):
|
||||
build_guided_decoding_logits_processor_async(
|
||||
sampling_params=params,
|
||||
tokenizer=await self.get_tokenizer(lora_request),
|
||||
default_guided_backend=(self.decoding_config.guided_decoding_backend
|
||||
default_guided_backend=(self.decoding_config.backend
|
||||
if self.decoding_config
|
||||
else DecodingConfig.guided_decoding_backend),
|
||||
else DecodingConfig.backend),
|
||||
model_config=self.model_config,
|
||||
reasoning_backend=self.decoding_config.reasoning_backend,
|
||||
)
|
||||
|
||||
Reference in New Issue
Block a user