Skip to content

Pull requests: AMD-AGI/Hyperloom

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

ci: expose test progress and nonfatal hang tracebacks
#1492 opened Sep 11, 2026 by zengleixin-amd Contributor Loading…
Feat/vllm v29 patch
#1487 opened Sep 11, 2026 by devalshahamd Collaborator Loading…
Removing patching for SGLang versions
#1486 opened Sep 11, 2026 by mohbasit Collaborator Loading…
feat(baremetal): bump vLLM default to 0.28.0 with glibc gate
#1476 opened Sep 10, 2026 by BaoYunkai Contributor Loading…
3 tasks done
fix(roofline): drop only cuda-graph capture, not inductor, for vLLM
#1470 opened Sep 10, 2026 by mgehre-amd Collaborator Loading…
Stop reporting unknown GPU archs as MI300X (baremetal)
#1460 opened Sep 9, 2026 by mgehre-amd Collaborator Loading…
feat(trace): per-LLM-call attribution, and the reports that read it domain:inference Related to inference optimization type:feature New capability or improvement
#1450 opened Sep 8, 2026 by rodosingh Member Loading…
ProTip! What’s not been updated in a month: updated:<2026-08-12.