Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
900 Open 10,094 Closed
900 Open 10,094 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

model : refactor bias tensor names model Model specific refactoring Refactoring
#22079 opened Apr 18, 2026 by CISC Member Loading…
[SYCL] Update oneapi 2025.3.3, Seperate SYCL build, release Ubuntu 24 package. devops improvements to build systems and github actions documentation Improvements or additions to documentation SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#22078 opened Apr 18, 2026 by NeoZhangJianyu Contributor Loading…
1
common/autoparser : allow space after tool call testing Everything test related
#22073 opened Apr 18, 2026 by aldehir Contributor Loading…
1
sycl: Battlemage (BMG) optimizations — AOT, Q5_K reorder, PAD stride fix, new ops, oneMKL routing ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#22066 opened Apr 17, 2026 by aicss-genai Loading…
2
Extend LoRA hotswapping support examples python python script changes server
#22061 opened Apr 17, 2026 by skiz Loading…
1
GGML: Allow static build with dynamic loaded backends ggml changes relating to the ggml tensor library for machine learning
#22059 opened Apr 17, 2026 by ervanalb Loading…
2 tasks done
spec: save the dynamic/static ngram cache file
#22055 opened Apr 17, 2026 by petersid2022 • Draft
quant: handle shared-KV layer tensors in imatrix-dependent quantization testing Everything test related
#22054 opened Apr 17, 2026 by ajfonthemove Loading…
3 tasks
2
CUDA: refactor mma data loading for AMD ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#22051 opened Apr 17, 2026 by JohannesGaessler Contributor Loading…
ggml-webgpu: reset CPU/GPU profiling time when freeing context ggml changes relating to the ggml tensor library for machine learning WebGPU
#22050 opened Apr 17, 2026 by yomaytk Contributor Loading…
1
2
Reduce CPU overhead in meta backend: cache subgraph splits when cgraph is unchanged ggml changes relating to the ggml tensor library for machine learning
#22041 opened Apr 17, 2026 by gaugarg-nv Contributor Loading…
8
server: Skip API key verification for static files examples server
#22038 opened Apr 17, 2026 by roj234 Contributor Loading…
mtmd, llama : Update HunyuanVL vision-language model support examples model Model specific python python script changes
#22037 opened Apr 17, 2026 by ManaEstras Loading…
3 tasks done
1
[SYCL] Fix reorder MMVQ assert on unaligned vocab sizes ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#22035 opened Apr 17, 2026 by PMZFX Contributor Loading…
1
4
server: log prompts to directory examples server
#22031 opened Apr 17, 2026 by jacekpoplawski Contributor • Draft
1
mtmd, llama, ggml : Update HunyuanVL support Apple Metal https://en.wikipedia.org/wiki/Metal_(API) examples ggml changes relating to the ggml tensor library for machine learning model Model specific Nvidia GPU Issues specific to Nvidia GPUs OpenCL Issues specific to the OpenCL backend python python script changes SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language testing Everything test related Vulkan Issues specific to the Vulkan backend
#22029 opened Apr 17, 2026 by ManaEstras Loading…
3 tasks done
10
server: Expose media_marker on /props endpoint. examples server
#22028 opened Apr 17, 2026 by cetarthoriphros Loading…
llama-mmap: add MADV_HUGEPAGE hint for THP on Linux
#22022 opened Apr 16, 2026 by Marxist-Leninist Contributor Loading…
ggml-vulkan/CMakeLists: add a check for SPIRV-Headers ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#22009 opened Apr 16, 2026 by jeeb • Draft
sampling: add segment-level repetition loop detection examples server
#22007 opened Apr 16, 2026 by Frank-Schruefer Loading…
12
opencl: workaround Adreno LLVM compiler SIGSEGV in subgroup arithmetic ops ggml changes relating to the ggml tensor library for machine learning OpenCL Issues specific to the OpenCL backend
#22006 opened Apr 16, 2026 by RokketCrypto Loading…
1
model: move load_hparams and load_tensors to per-model definition model Model specific python python script changes
#22004 opened Apr 16, 2026 by ngxson Contributor Loading…
4 of 6 tasks
1
@ggerganov
19
server : fix proxy client timeout in router mode examples server
#22003 opened Apr 16, 2026 by xris99 Loading…
1
8
server: Enable transcriptions API for LFM2-Audio examples server
#22000 opened Apr 16, 2026 by tdakhran Contributor Loading…
5
rpc : refactor the RPC transport ggml changes relating to the ggml tensor library for machine learning merge ready A maintainer can use this label to indicate that they consider the changes final and ready to merge.
#21998 opened Apr 16, 2026 by rgerganov Member Loading…
@rgerganov
1
ProTip! Adding no:label will show everything without a label.