qsa-hisparse
QSA HiSparse as an SGLang plugin: 256K CPU KV offload and text/image host prefix reuse for Qwen Sparse Attention on unmodified SGLang
- 3
- Python
- Apache-2.0
- 2 wk ago
QSA HiSparse as an SGLang plugin: 256K CPU KV offload and text/image host prefix reuse for Qwen Sparse Attention on unmodified SGLang