Skip to content

Actions: sgl-project/sglang

Actions

PR Test (AMD)

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
pr-test-amd.yml
31,719 workflow runs
31,719 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[HiCache] support memory_pool_host page head layout PR Test (AMD) #38263: Pull request #11644 labeled by huangtingwei9988
Queued antgroup:support_page_head_layout
antgroup:support_page_head_layout
Queued
[VLM] Support piecewise cuda graph for vlm PR Test (AMD) #38262: Pull request #13055 synchronize by yuan-luo
15s antgroup:support_vl_pcg
antgroup:support_vl_pcg
15s
[VLM] Support piecewise cuda graph for vlm PR Test (AMD) #38261: Pull request #13055 synchronize by yuan-luo
15s antgroup:support_vl_pcg
antgroup:support_vl_pcg
15s
router select dp group with the minimum number of tokens PR Test (AMD) #38260: Pull request #13208 synchronize by jiashaokun-1
13s jiashaokun-1:main
jiashaokun-1:main
13s
Opt tp: tp attn support tp reduce scattered input (#10568) PR Test (AMD) #38259: Commit d91b16e pushed by ch-wan
Queued main
main
Queued
[wip] [11/N] MoE Refactor: Simplifying SBO Implementation with Dispatcher Hooks PR Test (AMD) #38258: Pull request #13327 synchronize by ch-wan
Queued cheng/refactor/sbo
cheng/refactor/sbo
Queued
[wip] [11/N] MoE Refactor: Simplifying SBO Implementation with Dispatcher Hooks PR Test (AMD) #38257: Pull request #13327 labeled by sglang-bot
3m 45s cheng/refactor/sbo
cheng/refactor/sbo
3m 45s
[VLM] Support piecewise cuda graph for vlm PR Test (AMD) #38256: Pull request #13055 synchronize by yuan-luo
16s antgroup:support_vl_pcg
antgroup:support_vl_pcg
16s
Fix: add missing get_embed_and_head in MiniMax M2 for Eagle3 (#13297) PR Test (AMD) #38255: Commit b051d76 pushed by zhyncs
32m 15s main
main
32m 15s
Super tiny expose transform_scale_ue8m0 API for RL frameworks (#13323) PR Test (AMD) #38254: Commit d971f22 pushed by fzyzcjy
5m 6s main
main
5m 6s
[bug fix] split the images of one request into multiparts PR Test (AMD) #38253: Pull request #11828 synchronize by stmatengss
Queued openanolis:Xuchun/mm-dev
openanolis:Xuchun/mm-dev
Queued
chore: bump sgl-kernel version to 0.3.17.post1 PR Test (AMD) #38252: Pull request #13325 labeled by sglang-bot
Queued bot/bump-kernel-version-0.3.17.post1-e9aa
bot/bump-kernel-version-0.3.17.post1-e9aa
Queued
Add FP32 dtype support for RoPE - Part1 PR Test (AMD) #38251: Pull request #13181 synchronize by jinyouzhi
Queued jinyouzhi:rope_fp32
jinyouzhi:rope_fp32
Queued
[opt kimi k2 1 / n] Add kimi k2 moe fused gate (#13287) PR Test (AMD) #38250: Commit 1d3d42b pushed by ispobock
16m 52s main
main
16m 52s
Update marlin moe kernel interface (#13322) PR Test (AMD) #38249: Commit 8e9f05e pushed by ispobock
3m 48s main
main
3m 48s
Support weight update for blackwell DeepGEMM PR Test (AMD) #38248: Pull request #13324 labeled by sglang-bot
Queued fzyzcjy:feat/inv_trans_scale
fzyzcjy:feat/inv_trans_scale
Queued
Super tiny expose transform_scale_ue8m0 API for RL frameworks PR Test (AMD) #38247: Pull request #13323 labeled by sglang-bot
28m 13s fzyzcjy:feat/transform_scale_ue8m0
fzyzcjy:feat/transform_scale_ue8m0
28m 13s
[RL] support update_weights_from_tensor for mtp (#7415) PR Test (AMD) #38246: Commit bc08352 pushed by fzyzcjy
13m 43s main
main
13m 43s
Update marlin moe kernel interface PR Test (AMD) #38245: Pull request #13322 synchronize by ispobock
17m 22s ke/marlin-moe-interface
ke/marlin-moe-interface
17m 22s
[NVIDIA] Fix broken fp8 MoE of deepseek v3 PR Test (AMD) #38244: Pull request #13264 synchronize by kaixih
Queued kaixih:fix_broken_fp8_moe
kaixih:fix_broken_fp8_moe
Queued
Tiny refactor condition to requant scale ue8m0 (#13286) PR Test (AMD) #38243: Commit 33f08a9 pushed by fzyzcjy
21m 11s main
main
21m 11s
Tiny refactor condition to requant scale ue8m0 PR Test (AMD) #38242: Pull request #13286 synchronize by fzyzcjy
31s fzyzcjy:feat/refactor_requant
fzyzcjy:feat/refactor_requant
31s
Support inverse transform ue8m0 scale (#13285) PR Test (AMD) #38241: Commit 8e6083b pushed by fzyzcjy
1m 39s main
main
1m 39s
Support fast gemm when in batch invariant DeepGEMM fallback (#13259) PR Test (AMD) #38240: Commit 2fbc78a pushed by fzyzcjy
22s main
main
22s
[RL] Allow bypassing /health check (#13320) PR Test (AMD) #38239: Commit f0b5ccf pushed by fzyzcjy
26s main
main
26s