Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
524 Open 6,853 Closed
524 Open 6,853 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

kleidiai: generalize compute_forward_kv_cache to compute_forward_fp16
#15817 opened Sep 5, 2025 by chaxu01 Loading…
ggml : split graph allocations according to backend max buffer size ggml changes relating to the ggml tensor library for machine learning testing Everything test related Vulkan Issues specific to the Vulkan backend
#15815 opened Sep 5, 2025 by Acly Loading…
3
CANN: implement LRU cache for ACL graphs in CANN backend Ascend NPU issues specific to Ascend NPUs ggml changes relating to the ggml tensor library for machine learning
#15814 opened Sep 5, 2025 by noemotiovon Loading…
CUDA: Conv2d Tensor Core ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#15813 opened Sep 5, 2025 by mnehete32 Loading…
5
CANN: Switch to stream synchronization Ascend NPU issues specific to Ascend NPUs ggml changes relating to the ggml tensor library for machine learning
#15809 opened Sep 5, 2025 by noemotiovon Loading…
Add conv2d Implicit GEMM ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs testing Everything test related
#15805 opened Sep 4, 2025 by bssrdf • Draft
20
CUDA: fastdiv, launch bounds for mmvq + q8_1 quant ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#15802 opened Sep 4, 2025 by JohannesGaessler Loading…
6
vulkan: add mul_mat variant for embedded gpus ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#15800 opened Sep 4, 2025 by rmatif • Draft
23
ggml-backend : add GGML_BACKEND_DEVICE_TYPE_IGPU device type examples ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs Vulkan Issues specific to the Vulkan backend
#15797 opened Sep 4, 2025 by slaren Loading…
vulkan: support im2col_3d ggml changes relating to the ggml tensor library for machine learning testing Everything test related Vulkan Issues specific to the Vulkan backend
#15795 opened Sep 4, 2025 by jeffbolznv Loading…
vulkan: Support pad_ext ggml changes relating to the ggml tensor library for machine learning testing Everything test related Vulkan Issues specific to the Vulkan backend
#15794 opened Sep 4, 2025 by jeffbolznv Loading…
Implement --log-colors with always/never/auto
#15792 opened Sep 4, 2025 by ericcurtin Loading…
7
Add docker:// protocol support for llama-server model pulling
#15790 opened Sep 4, 2025 by ericcurtin Loading…
35
lama devops improvements to build systems and github actions
#15788 opened Sep 4, 2025 by Karen86Tonoyan Loading…
Add model to header title with mouseover examples server
#15787 opened Sep 4, 2025 by pudepiedj Loading…
docker : Fix AMDGPU_TARGETS deprecated warnning message devops improvements to build systems and github actions
#15786 opened Sep 4, 2025 by haiyuewa Loading…
ggml: allow casting between f32 and i32 Apple Metal https://en.wikipedia.org/wiki/Metal_(API) ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs testing Everything test related Vulkan Issues specific to the Vulkan backend
#15783 opened Sep 4, 2025 by ngxson Loading…
4 tasks done
3
example: add prediction-next-token command line argument handling Example for show probability of next token examples
#15774 opened Sep 3, 2025 by no4ni Loading…
CUDA: faster tile FA (Pascal/AMD), headsize 256 ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#15769 opened Sep 3, 2025 by JohannesGaessler Loading…
1
CUDA: Add mul_mat_id support for the mmf kernel ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs testing Everything test related
#15767 opened Sep 3, 2025 by am17an Loading…
16
Consolidate multiple tensor copies to reduce API overhead ggml changes relating to the ggml tensor library for machine learning
#15750 opened Sep 2, 2025 by agray3 Loading…
1
1
nix: Added missing packages and options for ROCm build devops improvements to build systems and github actions ggml changes relating to the ggml tensor library for machine learning nix Issues specific to consuming flake.nix, or generally concerned with ❄ Nix-based llama.cpp deployment
#15747 opened Sep 2, 2025 by SteelPh0enix Loading…
ggml-cpu: fixes instability in NNPA Vector Intrinsics documentation Improvements or additions to documentation ggml changes relating to the ggml tensor library for machine learning
#15739 opened Sep 2, 2025 by taronaeo • Draft
2
2
Add scale_diag_mask_inf_softmax operation for transformer attention ggml changes relating to the ggml tensor library for machine learning testing Everything test related
#15738 opened Sep 2, 2025 by Arya-Hari • Draft
7
opencl: initial q8_0 mv support ggml changes relating to the ggml tensor library for machine learning OpenCL Issues specific to the OpenCL backend
#15732 opened Sep 2, 2025 by lhez Loading…
ProTip! Mix and match filters to narrow down what you’re looking for.