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 4, 11:15 UTC
Fix typo in log message for indexer cache (#41419)
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Michael Goin committed
Apr 30, 2026 at 18:02 UTC
75a4c166f25f8407591328d7ed92aa2231b0b841
1 file changed
+1
-1
vllm/model_executor/layers/deepseek_v4_attention.py
+1
-1
@@ -1052,7 +1052,7 @@ class DeepseekV4Indexer(nn.Module):
1052
self.compress_ratio = compress_ratio
1053
self.use_fp4_kv = self.vllm_config.attention_config.use_fp4_indexer_cache
1054
logger.info_once(
1055
- "Using %s indexer cache for Lighening Indexer.",
1055
+ "Using %s indexer cache for Lightning Indexer.",
1056
"MXFP4" if self.use_fp4_kv else "FP8",
1057
)
1058