Unverified50% confidenceFactExact time
By default, vLLM plans 90% of GPU VRAM into the budget via the --gpu-memory-utilization parameter, and only the remainder is used as KV cache.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/13/2026
First Seen
Valid until: 1/11/2027
Sources
Related Entities
Related Claims
UnverifiedvLLM的PagedAttention将KV缓存切分为固定大小物理块（通常16个token），使显存利用率从不足40%提升至接近90%76% similarUnverifiedAfter vLLM starts, it occupies more than 90% of VRAM to build a cache pool in pursuit of maximum speed.75% similarVerifiedPagedAttention splits the KV Cache into fixed-size physical blocks mapped via a block table, and can improve GPU utilization from below 40% to above 90%.74% similarUnverified传统LLM推理中，KV Cache的内存管理导致60%-80%的GPU显存被浪费72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1019927API
curl https://kongchang.com/api/v1/knowledge/claims/1019927MCP
get_claim(id=1019927)