Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
1,552 Open 13,251 Closed
1,552 Open 13,251 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

chat : preserve object alternatives in Qwen XML tool schemas documentation Improvements or additions to documentation testing Everything test related
#28651 opened Sep 9, 2026 by arc-uri-el Loading…
py : bump numpy to 2.4.6 conversion server
#28649 opened Sep 9, 2026 by CISC Member Loading…
1
webui: stop re-probing disabled /tools endpoint on every message server/ui
#28646 opened Sep 9, 2026 by geckguy Contributor Loading…
1 task done
1
metal: fix tuning include path in subprojects examples
#28644 opened Sep 9, 2026 by obrr-hhx • Draft
1
1
model: fix all granite family parameter counts model Model specific
#28643 opened Sep 9, 2026 by taronaeo Member Loading…
4
ggml-cpu: add Q4_0 8x8 gemv/gemm for riscv vlenb=16 case ggml changes relating to the ggml tensor library for machine learning
#28642 opened Sep 9, 2026 by hongyang-7 Loading…
2
common : reject non-positive batch size (#28525) testing Everything test related
#28639 opened Sep 9, 2026 by goodruyas Loading…
1
OpenVINO: optimize stateful decode and GPU MoE inference documentation Improvements or additions to documentation ggml changes relating to the ggml tensor library for machine learning OpenVINO
#28638 opened Sep 9, 2026 by wine99 Contributor Loading…
Convert: Support conversion to gguf for compressed-tensor mixed-precision NVFP4 checkpoint conversion
#28636 opened Sep 9, 2026 by ynankani Contributor • Draft
model : fix MTP context kv cache allocation for deepseek2, glm4moe, c… testing Everything test related
#28630 opened Sep 9, 2026 by LoganChu Loading…
1
1
hexagon: rope updates ggml changes relating to the ggml tensor library for machine learning Hexagon testing Everything test related
#28628 opened Sep 9, 2026 by tboinovski1 Contributor Loading…
Qwen4 next flash multi gpu & buffer size issues solved CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning model Model specific testing Everything test related
#28623 opened Sep 9, 2026 by gopinath87607 • Draft
1
ui : Accept WEBM video files server/ui
#28622 opened Sep 9, 2026 by EpicEric Loading…
1
cmake : Use x86-64 microarchitecture level -march when sufficient features are selected ggml changes relating to the ggml tensor library for machine learning
#28621 opened Sep 9, 2026 by bberberov Contributor Loading…
vulkan: use CPU writes in ggml_backend_vk_cpy_tensor_async if the context is idle ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#28618 opened Sep 8, 2026 by jeffbolznv Contributor Loading…
HIP: branch-free SWAR for __vsub4 / __vcmpne4 / __vcmpeq4 CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning
#28616 opened Sep 8, 2026 by SimonTeixidor Contributor • Draft
HIP: tune MMVQ batch thresholds on RDNA3.5 CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning
#28613 opened Sep 8, 2026 by SimonTeixidor Contributor Loading…
chat : keep DSML markup out of DeepSeek V3.2 string tool arguments testing Everything test related
#28612 opened Sep 8, 2026 by iamganesha • Draft
2
ggml-vulkan : tune L-tile warp micro-dimension for RDNA3 iGPUs ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#28611 opened Sep 8, 2026 by Troncooooo Loading…
1
mtmd: add KaniTTS-2 support conversion documentation Improvements or additions to documentation model Model specific mtmd Related to multimodal functionality (video/image/audio)
#28609 opened Sep 8, 2026 by hans00 • Draft
1
mtmd : add Soprano audio generation conversion documentation Improvements or additions to documentation mtmd Related to multimodal functionality (video/image/audio)
#28607 opened Sep 8, 2026 by hans00 Loading…
ggml-cpu(s390x): add Q1_0 vector intrinsic support documentation Improvements or additions to documentation 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.
#28606 opened Sep 8, 2026 by taronaeo Member Loading…
2
metal : fix MiniCPM3 crash with automatic flash attention Apple Metal https://en.wikipedia.org/wiki/Metal_(API) ggml changes relating to the ggml tensor library for machine learning testing Everything test related
#28599 opened Sep 8, 2026 by wyanzhao Contributor Loading…
1
json-schema-to-grammar : report incomplete-conversion warnings only once
#28598 opened Sep 8, 2026 by RIVOIRA Loading…
1
quantize: filter per-layer metadata when pruning layers (#27976) testing Everything test related
#28597 opened Sep 8, 2026 by Sumire-no-kai Loading…
1
1
ProTip! Exclude everything labeled bug with -label:bug.