-
Notifications
You must be signed in to change notification settings - Fork 47
Pull requests: AMD-AGI/Hyperloom
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Added support for ATOM detailed profiling
#1509
opened Sep 14, 2026 by
devalshahamd
Collaborator
Loading…
fix(provenance): record AITER installed under either distribution name
#1508
opened Sep 14, 2026 by
rpoornac
Collaborator
Loading…
Detect TheRock ROCm SDK wheel layout for LD_LIBRARY_PATH
#1503
opened Sep 14, 2026 by
lishuoshuo-amd
Contributor
Loading…
ci(forge-e2e): dispatch through Dispatron, keep the reporting here
#1496
opened Sep 12, 2026 by
zoroyihan7
Contributor
•
Draft
fix(kb): guard warm-start replay on the shape the canonical_id cannot express
#1495
opened Sep 11, 2026 by
rpoornac
Collaborator
Loading…
feat(explore): route EXPLORE (and SWEEP) through InferaSim instead of booting every candidate
#1494
opened Sep 11, 2026 by
araina-amd
Collaborator
Loading…
6 tasks
ci: expose test progress and nonfatal hang tracebacks
#1492
opened Sep 11, 2026 by
zengleixin-amd
Contributor
Loading…
Relocate the enablement subsystem and make ENABLEMENT the sixth phase
#1491
opened Sep 11, 2026 by
ZhengGong-amd
Collaborator
Loading…
feat(conc-sweep): grade the ladder on the axis the session keeps on, and report the guard beside it
#1490
opened Sep 11, 2026 by
haishuok0525
Collaborator
Loading…
6 tasks done
ci(e2e): dispatch through Dispatron, keep the reporting here
#1489
opened Sep 11, 2026 by
zoroyihan7
Contributor
•
Draft
Give the Radeon 8060S a board identity instead of unknown_gpu
#1483
opened Sep 10, 2026 by
mgehre-amd
Collaborator
•
Draft
Migrate
hyperloom-workload-optimizer skill to AMD-AGI/Hyperloom product repo
#1482
opened Sep 10, 2026 by
danielholanda
Loading…
feat(sbd-v6): declare the axis a session graded on, and record it where it is resolved
#1480
opened Sep 10, 2026 by
haishuok0525
Collaborator
Loading…
fix(roofline): drop only cuda-graph capture, not inductor, for vLLM
#1470
opened Sep 10, 2026 by
mgehre-amd
Collaborator
Loading…
【WIP】refactor(kernel): one session view, one lane shape, one integration entry
#1465
opened Sep 10, 2026 by
jiaqiang-dot-liu
Collaborator
•
Draft
2 of 3 tasks
fix(bench): stop baseline benchmarking from a private InferenceX copy
#1456
opened Sep 9, 2026 by
fengshaoyi-amd
Contributor
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…
fix(ci-e2e): budget queueing apart from running, and stop swallowing failed cancels
#1449
opened Sep 8, 2026 by
zoroyihan7
Contributor
Loading…
feat(forge): optimize compiler-emitted AMDGPU assembly
#1426
opened Sep 7, 2026 by
chennyiiis
Contributor
•
Draft
feat(atom): support Atom end to end — reap its recovery workers, default its kernel backend to forge, and publish it as supported
#1416
opened Sep 7, 2026 by
zengleixin-amd
Contributor
•
Draft
enablement: make the recipe's replay sufficiency fail closed on what a consumer actually receives
retest
Re-run E2E smoke without a new commit (runs once)
#1414
opened Sep 7, 2026 by
zoroyihan7
Contributor
•
Draft
docs(inference_optimizer): detach the optimizer the way the harness understands
#1400
opened Sep 3, 2026 by
zoroyihan7
Contributor
Loading…
fix(profile): stop the per-batch-size graph-capture exporter taking the server down
#1399
opened Sep 3, 2026 by
fengshaoyi-amd
Contributor
Loading…
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.