Skip to content

Pull requests: ROCm/ATOM

New pull request New
134 Open 1,043 Closed
134 Open 1,043 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

Route prefix-cache-hit prefill through sink ASM MHA kernel
#1345 opened Jun 24, 2026 by yhl-amd Contributor Loading…
1 task
[minimax_m3] gate AR+RMSNorm fusion on ATOM_ENABLE_ALLREDUCE_RMSNORM_…
#1344 opened Jun 24, 2026 by zejunchen-zejun Collaborator • Draft
1 task
Enable MiniMax-M3 vLLM plugin path
#1342 opened Jun 24, 2026 by lirui927 Contributor Loading…
1 task
1
feat: add backend for moe
#1341 opened Jun 24, 2026 by JiaoliangYu Contributor • Draft
1 task
Fix GC pause during CUDAGraph capture to prevent abort
#1339 opened Jun 24, 2026 by Jasen2201 Contributor Loading…
1 task
2
[gfx1151] Online INT8 W8A8 for Qwen3.6 27B / 35B-A3B on RDNA3.5, with working MTP
#1337 opened Jun 24, 2026 by carlushuang Collaborator Loading…
Feat/low bit ep
#1336 opened Jun 24, 2026 by JiaoliangYu Contributor • Draft
1 task
[MiniMax M3] Enable and Optimize the MiniMax M3 Eagle
#1333 opened Jun 24, 2026 by zejunchen-zejun Collaborator • Draft
2
[atom][m3] enable m3 tbo
#1324 opened Jun 23, 2026 by zhuyuhua-v Collaborator • Draft
1 task
feat(openai): Qwen3 (qwen3_coder/qwen3_xml) tool-call support
#1319 opened Jun 23, 2026 by carlushuang Collaborator Loading…
[Feature] OFFLOAD: standalone LMCache CPU/NVMe KV-offload connector
#1318 opened Jun 23, 2026 by yhl-amd Contributor Loading…
4 of 5 tasks
Add MiniMax-M3 (MXFP4/AttnFP8) model support
#1317 opened Jun 22, 2026 by thpereir Contributor Loading…
1 task
[KV-events] block token_offset + sequence numbers + replay
#1316 opened Jun 22, 2026 by bongwoobak Contributor Loading…
5
[gfx1151] Qwen3.5/3.6 (GDN hybrid) BF16 on RDNA3.5 via native Triton attention
#1314 opened Jun 22, 2026 by carlushuang Collaborator Loading…
[fea](qwen): support model runner v2 on qwen-next (#1249)
#1310 opened Jun 22, 2026 by PerryZhang01 Contributor Loading…
fix(minimax): restore qkv<=256 shape guard + harden compile-cache frozen-path guard
#1300 opened Jun 20, 2026 by sunway513 Collaborator Loading…
2
[Triton] [GFX12] Add gluon pa_decode_sparse from AITER
#1290 opened Jun 19, 2026 by k50112113 Collaborator Loading…
Feat: RTP-LLM plugin GLM5-FP8 support
#1289 opened Jun 18, 2026 by zhaoan12-prc Contributor Loading…
18
wip
#1284 opened Jun 18, 2026 by haoyangli0109 Contributor Loading…
[vllm-atom] support eagle3 for M3
#1283 opened Jun 18, 2026 by whx-sjtu Contributor • Draft
merge build sparse block into topk and enable fp8 kv cache
#1282 opened Jun 18, 2026 by ganyi1996ppo Contributor Loading…
1 task
Add mxfp8 x mxfp4 Triton MoE for DSv4
#1277 opened Jun 18, 2026 by azaidy Contributor Loading…
Fpz/mixed mla dispatch
#1271 opened Jun 18, 2026 by jiayyu Contributor • Draft
1 task
[atom][M3] add gemma+allreduce fusion for m3
#1261 opened Jun 17, 2026 by zhuyuhua-v Collaborator • Draft
1 task
[ci][mesh] add model P/D benchmark workflow
#1244 opened Jun 17, 2026 by wanzhenchn Contributor Loading…
ProTip! Follow long discussions with comments:>50.