Skip to content

Pull requests: ROCm/ATOM

New pull request New
151 Open 1,223 Closed
151 Open 1,223 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

wire GUGU - act+quant fusion into triton decode
#1570 opened Jul 12, 2026 by Boss2002n Contributor Loading…
5
[ATOM SGL] Minimax M3 FP8 & EAGLE
#1564 opened Jul 11, 2026 by ZhiweiYan-96 Contributor • Draft
1 task
1
[Frontend] DeepSeek-V4 native OpenAI/Anthropic/Responses API + DSML tool parser
#1563 opened Jul 10, 2026 by yhl-amd Contributor • Draft
3 tasks
1
[Frontend] Abort engine request on client disconnect (free leaked KV)
#1562 opened Jul 10, 2026 by yhl-amd Contributor Loading…
2 tasks
1
fix
#1559 opened Jul 10, 2026 by gbyu-amd Contributor • Draft
1 task
[feat][PCP] Enable PCP MoE-Merge+TBO for DeepSeek V4
#1553 opened Jul 10, 2026 by yitingw1 Contributor Loading…
1 task
3
[Feature] support Chunked Pipeline Parallelism + PD Disaggregation + Atomesh
#1552 opened Jul 10, 2026 by Jasen2201 Contributor • Draft
1 task
6
[sglang+atom] Fix radix-cache crash on MiniMax-M3
#1551 opened Jul 10, 2026 by ningding01 Contributor Loading…
1 task
1
fix gfx950 only
#1550 opened Jul 10, 2026 by feifei14119 Contributor Loading…
1 task
review-pr skill v2: dispatch gap rule, arch-constant FP fix, P5 timing error
#1549 opened Jul 10, 2026 by zufayu Contributor Loading…
5 tasks done
1
[atom-vllm] enable MTP for glm 5.2
#1531 opened Jul 9, 2026 by whx-sjtu Contributor • Draft
[ATOM-SGL/VLLM-CI] Modify testing cases for SGLang and vLLM.
#1530 opened Jul 9, 2026 by junyyang-amd Contributor Loading…
CI: change SGLang plugin M3 nightly accuracy threshold
#1529 opened Jul 9, 2026 by Phi-C Contributor Loading…
1 task
2
di ci: glm-5-2 1p 1d && 2p1d
#1528 opened Jul 9, 2026 by JiaoliangYu Contributor Loading…
1 task
[vLLM-ATOM] Support EAGLE3 spec decoding for MiniMax-M3
#1518 opened Jul 8, 2026 by kliuae Contributor • Draft
1 task
enable pcp for dsa models
#1514 opened Jul 8, 2026 by whx-sjtu Contributor • Draft
3
Enable GFX12 Preshuffle Weights
#1504 opened Jul 7, 2026 by amirumoAMD Contributor • Draft
1 task done
1
[Feature] Enable cache aware DP routing for standalone ATOMesh
#1501 opened Jul 7, 2026 by simondanielsson Loading…
1 task done
2
[feature] online quantize weights when loading weights
#1500 opened Jul 7, 2026 by haoyangli0109 Contributor • Draft
[KVConnector] native scale-up KV connector (HIP VMM, kv_connector=native)
#1499 opened Jul 7, 2026 by carlushuang Collaborator Loading…
2
fix(kv): reserve prefill-activation headroom and size KV pool for the tightest TP rank
#1495 opened Jul 7, 2026 by carlushuang Collaborator • Draft
feat(prezero): wire split-K GEMM prezero into Kimi MLA/MoE decode via a torch.compile pass
#1490 opened Jul 6, 2026 by ColorsWind Loading…
2
fix(moe): auto-degrade FP8 block_n/k from 128 to 64 on alignment mismatch (#442)
#1488 opened Jul 6, 2026 by haowu1234 Loading…
Optimize GLM5.2 MTP
#1481 opened Jul 6, 2026 by whx-sjtu Contributor Loading…
feat(quant): implement fp4_act_quant Triton kernel for DeepSeek-V4 (#807)
#1479 opened Jul 6, 2026 by haowu1234 Loading…
1
1
ProTip! Follow long discussions with comments:>50.