Skip to content

Pull requests: ROCm/llama.cpp

New pull request New
11 Open 49 Closed
11 Open 49 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

mmq: opt-in compacted MoE tiling for RDNA3.5
#63 opened Jul 21, 2026 by roberteg16 • Draft
2 of 3 tasks
tests: add MoE MMQ benchmark with routing-distribution generator
#62 opened Jul 20, 2026 by roberteg16 • Draft
3 tasks
ggml-cuda: add dequant-float matvec (mmvdq) for Q4_K/Q5_K/Q6_K on RDN…
#61 opened Jul 20, 2026 by Annieren Loading…
1
gfx1151 (Strix Halo): fuse attn_k+v into single MMVQ dispatch
#59 opened Jul 17, 2026 by jeffli-xilinx Loading…
7 tasks done
9
ggml-cuda: GEMM weight row padding + one-time K-padded f16 dequant for prefill
#57 opened Jul 17, 2026 by roberteg16 • Draft
4 of 5 tasks
CUDA: gated_delta_net - share per-token k/q via shared memory for long prefills
#54 opened Jul 16, 2026 by roberteg16 • Draft
1 of 2 tasks
1
ggml-cuda: use mma-f16 flash-attn for D=256 on RDNA3.5
#40 opened Jul 8, 2026 by roberteg16 • Draft
1
RDNA3.5 (gfx11 / gfx1151) MMQ prefill optimizations
#32 opened Jun 30, 2026 by liangliangchang Loading…
4 tasks done
5
Gfx11
#29 opened Jun 26, 2026 by liangliangchang • Draft
Basic CI for hrx-v2
#27 opened Jun 18, 2026 by AaronStGeorge Collaborator Loading…
Updates for new hrx C and CMake APIs
#26 opened Jun 18, 2026 by AaronStGeorge Collaborator Loading…
ProTip! Add no:assignee to see everything that’s not assigned.