Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
626 Open 8,220 Closed
626 Open 8,220 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

eval-callback : add support for saving logits examples
#18281 opened Dec 22, 2025 by danbev Loading…
2
Vulkan: Tune Flash Attention for MoE on AMD GPUs ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#18280 opened Dec 22, 2025 by 0cc4m Loading…
Prevent crash if TTFT >300sec, boosted to 90 days examples server
#18279 opened Dec 22, 2025 by wbtek Loading…
opencl: fix q4_0 unpacking for adreno in get_tensor ggml changes relating to the ggml tensor library for machine learning OpenCL Issues specific to the OpenCL backend
#18278 opened Dec 22, 2025 by lhez • Draft
tools : use common_log_pause to fix fit-params output race examples
#18276 opened Dec 22, 2025 by Aadeshveer Loading…
1
KYLIN: fix compile error for cuda backend ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#18275 opened Dec 22, 2025 by lizhenneng Loading…
1
docs: Fix typos in SYCL documentation documentation Improvements or additions to documentation SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#18269 opened Dec 21, 2025 by yoka Loading…
add LLAMA_ARG_OVERRIDE_TENSOR env var for -ot arg
#18267 opened Dec 21, 2025 by ddh0 Loading…
llama: fix magic number of 999 for GPU layers
#18266 opened Dec 21, 2025 by JohannesGaessler Loading…
1
2
server: add real-time prompt preprocessing progress via synthetic SSE chunks examples server
#18265 opened Dec 21, 2025 by ServeurpersoCom • Draft
1
3
vulkan: Extend rope fusions to allow mrope ggml changes relating to the ggml tensor library for machine learning testing Everything test related Vulkan Issues specific to the Vulkan backend
#18264 opened Dec 21, 2025 by jeffbolznv Loading…
3
server: prevent data race from HTTP threads examples server
#18263 opened Dec 21, 2025 by ngxson Loading…
1
16
server: (docs) remove mention about extra_args examples server
#18262 opened Dec 21, 2025 by ngxson Loading…
Add Gemma3n multimodal support with MobileNetV5 vision encoder examples model Model specific python python script changes
#18256 opened Dec 21, 2025 by simrnsingh Loading…
29
ggml rpc : Add missing check for rpc buffer type ggml changes relating to the ggml tensor library for machine learning
#18242 opened Dec 21, 2025 by struct Loading…
ggml-cpu: parallelize tensor repacking with OpenMP ggml changes relating to the ggml tensor library for machine learning
#18239 opened Dec 21, 2025 by pestopoppa Loading…
cli: buffering info log, only show if model load failed examples
#18236 opened Dec 20, 2025 by ngxson Loading…
8
webui: Fix the header backdrop blur examples server
#18230 opened Dec 20, 2025 by ImadSaddik Loading…
server: /v1/responses (text generation only) examples python python script changes server
#18227 opened Dec 20, 2025 by openingnow Loading…
6
webui: use server presets as parameter placeholders examples server
#18226 opened Dec 20, 2025 by ServeurpersoCom Loading…
1
ggml-metal: guard buffer map slicing Apple Metal https://en.wikipedia.org/wiki/Metal_(API) ggml changes relating to the ggml tensor library for machine learning
#18225 opened Dec 20, 2025 by SzymonPrajs Loading…
webui: apply webui_settings on first load examples server
#18223 opened Dec 20, 2025 by ServeurpersoCom Loading…
1
ggml-metal: fix memset range and temp buffer leaks Apple Metal https://en.wikipedia.org/wiki/Metal_(API) ggml changes relating to the ggml tensor library for machine learning
#18221 opened Dec 20, 2025 by SzymonPrajs Loading…
model: support nvidia/llama-embed-nemotron model Model specific python python script changes
#18220 opened Dec 20, 2025 by sfallah • Draft
convert: rework ftype heuristics python python script changes
#18214 opened Dec 20, 2025 by taronaeo Loading…
1
9
ProTip! no:milestone will show everything without a milestone.