@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / a822eb3413
Fork 0

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, 01:45 UTC

[Misc] fix typo in block manager (#4453)

Pastel! committed May 1, 2024 at 11:41 UTC a822eb3413087062a38cea495564ec4a7093c3e5
1 file changed +1 -1
vllm/core/block_manager_v1.py
+1 -1
@@ -391,7 +391,7 @@ class BlockSpaceManagerV1(BlockSpaceManager):
391 block_table.append(block_table[len(block_table) %
392 self.block_sliding_window])
393 else:
394 - # The sequence has a new logical block.
394 + # The sequence hash a new logical block.
395 # Allocate a new physical block.
396 new_block = self._allocate_last_physical_block(seq)
397 block_table.append(new_block)