Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
866 Open 9,928 Closed
866 Open 9,928 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

server: fix SWA prompt reuse boundary condition
#21695 opened Apr 9, 2026 by 1oridevs • Draft
3 tasks done
2
server : improve cache reuse diagnostics for SWA and hybrid models
#21693 opened Apr 9, 2026 by 1oridevs Loading…
2
debug: functionality to dump full tensors and compare examples python python script changes
#21691 opened Apr 9, 2026 by pwilkin Member Loading…
common: mark --split-mode tensor as experimental
#21684 opened Apr 9, 2026 by JohannesGaessler Contributor Loading…
cli: add option to connect to server via http(s) examples
#21674 opened Apr 9, 2026 by pwilkin Member • Draft
6
Bug-Fix sets an upper VRAM limit for cached ggml_cuda graphs to prevent VRAM memory leaks ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#21673 opened Apr 9, 2026 by kmorennv Loading…
4
webui: Fix messages rendering for "Show raw output" examples server/webui server
#21672 opened Apr 9, 2026 by allozaur Contributor Loading…
common : add fluidity to the progress bar
#21671 opened Apr 9, 2026 by angt Member Loading…
common : fix when loading a cached HF models with unavailable API
#21670 opened Apr 9, 2026 by angt Member Loading…
ggml-webgpu: support non-square subgroup matrix configs for Intel GPUs ggml changes relating to the ggml tensor library for machine learning WebGPU
#21669 opened Apr 9, 2026 by SharmaRithik Loading…
webui: Static build output improvements examples server/webui server
#21667 opened Apr 9, 2026 by allozaur Contributor Loading…
CUDA: fuse muls ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#21665 opened Apr 9, 2026 by am17an Contributor Loading…
2
Convert: Fix NemotronH Config Parsing python python script changes
#21664 opened Apr 9, 2026 by anavp-nvidia Contributor Loading…
1
common: load parser in common_chat_parser_params constructor
#21659 opened Apr 9, 2026 by sacredvoid Contributor Loading…
1
Fix WebUI thinking mode request handling
#21657 opened Apr 9, 2026 by redyuan43 Loading…
1
docker: add OCI image labels for version and build date devops improvements to build systems and github actions
#21653 opened Apr 9, 2026 by ssam18 Contributor Loading…
1
9
Prevent the sum of the dequantized activation in q8_1 from overflowing
#21652 opened Apr 9, 2026 by bartowski1182 Contributor Loading…
2
ci: add android arm64 build and release
#21647 opened Apr 8, 2026 by ykhrustalev Contributor Loading…
convert : force f16 or f32 on step3-vl conv weights
#21646 opened Apr 8, 2026 by CISC Member Loading…
ggml-webgpu: Update register tiling matmul to use f32 accumulation
#21644 opened Apr 8, 2026 by reeselevine Contributor Loading…
1
[SYCL] Fix Q8_0 reorder: garbage on 2nd prompt + crash on full VRAM ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#21638 opened Apr 8, 2026 by PMZFX Contributor Loading…
1
1
fix(openvino): define PartialShape bounds for tensors
#21637 opened Apr 8, 2026 by thedanhoffman Contributor Loading…
4
(Performance; ggml-cpu) Optimized x86 and generic cpu q1_0 dot (follow up)
#21636 opened Apr 8, 2026 by pl752 Contributor Loading…
1
Enable ccache on riscv64
#21632 opened Apr 8, 2026 by luhenry Contributor • Draft
7
cmake: fix CMP0194 warning on Windows with MSVC
#21630 opened Apr 8, 2026 by texasich Loading…
1
1
ProTip! Type g p on any issue or pull request to go back to the pull request listing page.