Unverified60% confidenceFactTime unknown
When the context length is extended from 4K to 8K, the VRAM occupied by the KV Cache nearly doubles
2
Sources
60%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
本地部署大模型怎么判断显存爆了？一文看懂显存监控方法
bilibili步步梅
Related Entities
Related Claims
VerifiedWhen the context length is extended to one million tokens, the KV Cache itself can occupy tens of gigabytes of VRAM.79% similarVerifiedLLaMA-2 7B's KV Cache occupies approximately 2 GB with a 4096-token context, and can surge to over 16 GB when extended to a 32K context75% similarVerifiedThe VRAM usage of KV Cache grows linearly with sequence length; for a 70-billion-parameter model using GQA, the KV Cache usage can reach tens of GB when processing 100,000 tokens.74% similarUnverifiedWhen the context expands from 4K tokens to 128K tokens (a 32× increase), the theoretical computation grows by approximately 1024×.74% similarUnverifiedK8V4非对称量化将Key压缩到8位、Value压缩到4位，使128K上下文的KV Cache从16GB降到约6GB72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/17566API
curl https://kongchang.com/api/v1/knowledge/claims/17566MCP
get_claim(id=17566)