0
代码
代码
Issues
Pull Requests
讨论
Wiki
分析
0
GGitHub[doc](cp) correct the prefill of GQA and adjust desc of block table. (#5697)
38cfcd57创建于 1月19日历史提交
文件最后提交记录最后更新时间
blocktable.png
[doc](cp) correct the prefill of GQA and adjust desc of block table. (#5697) ### What this PR does / why we need it? correct the seq length of KV for prefill of GQA and clarify the desc of block table distribution in developer guide. - vLLM version: v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d --------- Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
chunkedprefill.png
[doc]<PCP&DCP> add developer guide for PCP&DCP (#5372) ### What this PR does / why we need it? add developer guide for PCP&DCP - vLLM version: release/v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/bc0a5a0c089844b17cb93f3294348f411e523586 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
dcp-decode.png
[doc]<PCP&DCP> add developer guide for PCP&DCP (#5372) ### What this PR does / why we need it? add developer guide for PCP&DCP - vLLM version: release/v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/bc0a5a0c089844b17cb93f3294348f411e523586 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
dcp-prefill.png
[doc]<PCP&DCP> add developer guide for PCP&DCP (#5372) ### What this PR does / why we need it? add developer guide for PCP&DCP - vLLM version: release/v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/bc0a5a0c089844b17cb93f3294348f411e523586 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
device_world.png
[doc](cp) correct the prefill of GQA and adjust desc of block table. (#5697) ### What this PR does / why we need it? correct the seq length of KV for prefill of GQA and clarify the desc of block table distribution in developer guide. - vLLM version: v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d --------- Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
head-tail-style.png
[doc]<PCP&DCP> add developer guide for PCP&DCP (#5372) ### What this PR does / why we need it? add developer guide for PCP&DCP - vLLM version: release/v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/bc0a5a0c089844b17cb93f3294348f411e523586 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
overview.png
[doc]<PCP&DCP> add developer guide for PCP&DCP (#5372) ### What this PR does / why we need it? add developer guide for PCP&DCP - vLLM version: release/v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/bc0a5a0c089844b17cb93f3294348f411e523586 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
pcp-decode.png
[doc]<PCP&DCP> add developer guide for PCP&DCP (#5372) ### What this PR does / why we need it? add developer guide for PCP&DCP - vLLM version: release/v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/bc0a5a0c089844b17cb93f3294348f411e523586 Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前
pcp-prefill.png
[doc](cp) correct the prefill of GQA and adjust desc of block table. (#5697) ### What this PR does / why we need it? correct the seq length of KV for prefill of GQA and clarify the desc of block table distribution in developer guide. - vLLM version: v0.13.0 - vLLM main: https://github.com/vllm-project/vllm/commit/2f4e6548efec402b913ffddc8726230d9311948d --------- Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>4 个月前