Skip to content

Pull requests: ROCm/ATOM

New pull request New
125 Open 762 Closed
125 Open 762 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Remove Triton build stage from Dockerfile
#1035 opened Jun 2, 2026 by gyohuangxin Member Loading…
2
prefill gdr autotune envvar update
#1034 opened Jun 2, 2026 by ganyi1996ppo Contributor Loading…
1 task
fix: chunk prefill
#1032 opened Jun 2, 2026 by jiayyu Contributor Loading…
1 task
1
use ATOM_USE_FP4_TRITON_GEMM to enable non shuffle triton gemm
#1031 opened Jun 2, 2026 by zhuyuhua-v Collaborator • Draft
1 task
1
Cherry pick aw benchmark
#1029 opened Jun 2, 2026 by junyyang-amd Contributor Loading…
[feature] Support deepseekv4 online quantization
#1027 opened Jun 2, 2026 by haoyangli0109 Contributor • Draft
Fix LMCache offload reload handoff
#1026 opened Jun 2, 2026 by yhl-amd Contributor • Draft
fix: warmup uses full token budget for DP
#1024 opened Jun 2, 2026 by ZhangLirong-amd Contributor Loading…
2
Add HIP MLA reduce kernel dispatch logic
#1018 opened Jun 1, 2026 by ftyghome Loading…
3
Enable DPA + TBO Support for Kimi K2.5 Inference
#1017 opened Jun 1, 2026 by ftyghome Loading…
9
[atom-sgl-benchmark] Release mi308 benchmark results to dashboard
#1008 opened Jun 1, 2026 by junyyang-amd Contributor Loading…
Gfx1250 bringup moe
#1007 opened Jun 1, 2026 by yadaish • Draft
1 task
[ATOM-vLLM] Upgrade vLLM version to v0.22.0
#1006 opened Jun 1, 2026 by whx-sjtu Contributor • Draft
1
[atom-vllm-benchmark] Change AW execution logic from one server one job to one server multi jobs
#1005 opened Jun 1, 2026 by junyyang-amd Contributor Loading…
2
[Fix] Fix sparse_attn_v4_paged_prefill for MI308
#1003 opened Jun 1, 2026 by yitingw1 Loading…
1 task
2
[kv_transfer] Add LMCache KV offload connector for ATOM
#997 opened Jun 1, 2026 by yhl-amd Contributor Loading…
feat: support DeepSeek-V4-Flash-Base model on gfx942 device.
#996 opened Jun 1, 2026 by junna2016 Loading…
20
ci(sglang): add Kimi-K2 e2e accuracy + perf regression
#985 opened May 30, 2026 by sunway513 Collaborator • Draft
[atom-sgl-benchmark] Debug benchmark timeout
#977 opened May 29, 2026 by junyyang-amd Contributor Loading…
[sgl][recipe]update sglang-atom ds fp4 recipe
#975 opened May 29, 2026 by zhuyuhua-v Collaborator • Draft
4
455 test yadai dump
#969 opened May 28, 2026 by yadaish • Draft
1 task
Enable MTP for v2 model
#964 opened May 28, 2026 by ZhiweiYan-96 Contributor • Draft
4
[Feat] Fused qknorm + quant for dpsk v2 model
#963 opened May 28, 2026 by qichu-yun Contributor Loading…
1 task
2
deepseek v4 fp8_einsum enable
#960 opened May 28, 2026 by ganyi1996ppo Contributor • Draft
1 task
1
[Feat] enable fuse share expert in DeepSeek-R1-0528-MXFP4-v2 in sgl atom
#958 opened May 28, 2026 by ZLkanyo009 Contributor Loading…
ProTip! Exclude everything labeled bug with -label:bug.