XiaoxinWang e38ef2c434 support FULL graph mode for GQA (#3970) ...

### What this PR does / why we need it?
The current library only supports the FullDecodeOnly graph mode, which
enables full graph execution during the decode. This PR extends support
to allow full graph execution in both the prefill and decode, referred
to as FULL graph mode.

- vLLM version: v0.11.0
- vLLM main:
2918c1b49c

Signed-off-by: wangxiaoxin-sherie <wangxiaoxin7@huawei.com>
Co-authored-by: wangxiaoxin-sherie <wangxiaoxin7@huawei.com>

2025-11-17 10:50:35 +08:00

..

ISSUE_TEMPLATE

[Doc] Release note for v0.11.0rc0 (#3224)

2025-09-30 03:26:18 +08:00

workflows

support FULL graph mode for GQA (#3970)

2025-11-17 10:50:35 +08:00

actionlint.yaml

[1/N][CI] Add multi node test (#3359)

2025-10-11 14:50:46 +08:00

dependabot.yml

[CI] Add dependabot support and labeler workflow (#162)

2025-02-27 19:46:31 +08:00

Dockerfile.buildwheel

Upgrade CANN to 8.3.rc1 (#3945)

2025-11-03 20:21:07 +08:00

Dockerfile.nightly.a2

[CI] Add daily images build for nightly ci (#3989)

2025-11-13 20:10:12 +08:00

Dockerfile.nightly.a3

[CI] Add daily images build for nightly ci (#3989)

2025-11-13 20:10:12 +08:00

format_pr_body.sh

Remove redundant empty lines in commit msg (#1814)

2025-07-16 16:50:44 +08:00

labeler.yml

[CI] Add dependabot support and labeler workflow (#162)

2025-02-27 19:46:31 +08:00

PULL_REQUEST_TEMPLATE.md

[1/2/N] Enable pymarkdown and python __init__ for lint system (#2011)

2025-07-25 22:16:10 +08:00