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 14, 05:30 UTC
[Bugfix] Fix parsing of `--disable-mm-preprocessor-cache` (#22909)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Cyrus Leung committed
Aug 14, 2025 at 23:09 UTC
dbe298046c8a2528c48cbd2ceca0d074052054c4
1 file changed
+1
-1
vllm/engine/arg_utils.py
+1
-1
@@ -711,7 +711,7 @@ class EngineArgs:
711
"--mm-processor-cache-gb",
712
**multimodal_kwargs["mm_processor_cache_gb"])
713
multimodal_group.add_argument("--disable-mm-preprocessor-cache",
714
- type=bool,
714
+ action="store_true",
715
deprecated=True)
716
multimodal_group.add_argument(
717
"--interleave-mm-strings",