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, 13:30 UTC
[Doc] Move todo out of beam search docstring (#17183)
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
Alex Brooks committed
Apr 25, 2025 at 05:44 UTC
7feae92c1fa44a979e0d449927a0282e73030f19
1 file changed
+2
-4
vllm/entrypoints/llm.py
+2
-4
@@ -519,11 +519,9 @@ class LLM:
519
prompts: A list of prompts. Each prompt can be a string or a list
520
of token IDs.
521
params: The beam search parameters.
522
-
523
- TODO: how does beam search work together with length penalty, frequency
524
- penalty, and stopping criteria, etc.?
522
"""
526
-
523
+ # TODO: how does beam search work together with length penalty,
524
+ # frequency, penalty, and stopping criteria, etc.?
525
beam_width = params.beam_width
526
max_tokens = params.max_tokens
527
temperature = params.temperature