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 17, 08:00 UTC
Only emit warning about internal tokenizer if it isn't being used (#939)
Nelson Liu committed
Sep 4, 2023 at 08:50 UTC
e15932bb60e645e533a4b2f999bec9c60328e6d3
1 file changed
+2
-1
vllm/transformers_utils/tokenizer.py
+2
-1
@@ -25,7 +25,8 @@ def get_tokenizer(
25
"Cannot use the fast tokenizer in slow tokenizer mode.")
26
kwargs["use_fast"] = False
27
28
- if "llama" in tokenizer_name.lower() and kwargs.get("use_fast", True):
28
+ if ("llama" in tokenizer_name.lower() and kwargs.get("use_fast", True)
29
+ and tokenizer_name != _FAST_LLAMA_TOKENIZER):
30
logger.info(
31
"For some LLaMA-based models, initializing the fast tokenizer may "
32
"take a long time. To eliminate the initialization time, consider "