A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 18:30 UTC
[Bugfix][Frontend] Fix validation of `logprobs` in `ChatCompletionRequest` (#14352)
Signed-off-by: Sebastian Schönnenbeck <sebastian.schoennenbeck@comma-soft.com>
Sebastian Schoennenbeck committed
Mar 18, 2025 at 13:50 UTC
dd732028f59dd149cfdd4e6ca899d50c5839051e
1 file changed
+1
-1
vllm/entrypoints/openai/protocol.py
+1
-1
@@ -548,7 +548,7 @@ class ChatCompletionRequest(OpenAIBaseModel):
548
if top_logprobs < 0:
549
raise ValueError("`top_logprobs` must be a positive value.")
550
551
- if not data.get("logprobs"):
551
+ if top_logprobs > 0 and not data.get("logprobs"):
552
raise ValueError(
553
"when using `top_logprobs`, `logprobs` must be set to true."
554
)