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 2, 02:15 UTC
[KV Offloading] Enable HMA models for Tiering Offloading (#44287)
Signed-off-by: varun sundar rabindranath <vsundarr@redhat.com> Co-authored-by: varun sundar rabindranath <vsundarr@redhat.com>
Varun Sundar Rabindranath committed
Jun 3, 2026 at 03:03 UTC
7268457999d54f4014f2815911837c9277ca23f5
1 file changed
-1
vllm/v1/kv_offload/tiering/spec.py
-1
@@ -110,7 +110,6 @@ class TieringOffloadingSpec(CPUOffloadingSpec):
110
self._scheduler_mmap = scheduler_mmap
111
112
# Create primary tier (CPU-based)
113
- assert len(self.gpu_block_size) == 1
113
primary_tier = CPUPrimaryTierOffloadingManager(
114
num_blocks=self.num_blocks,
115
cache_policy=self.eviction_policy, # type: ignore[arg-type]