Issue/168 InfiniLM接入paged attention接口#173

Open

PanZezhong1725 wants to merge 6 commits intomainfrom

Commits on Dec 30, 2025

issue/168 InfiniLM接入paged attention
PanZezhong1725
committed

Commits on Jan 4, 2026

issue/168 get contiguous paged kv cache
PanZezhong1725
committed

Commits on Jan 6, 2026

issue/168 remove input lengths
PanZezhong1725
committed

Commits on Jan 8, 2026

issue/168 support fixed paged attention api
PanZezhong1725
committed

Commits on Jan 9, 2026

issue/168 use n_blocks to init paged kv cache config, support fixed paged caching api
PanZezhong1725
committed

Commits on Jan 13, 2026

issue/168 fix bench.py
PanZezhong1725
committed