Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
1,230 Open 12,124 Closed
1,230 Open 12,124 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

tests : remove unnecessary sync in test-save-load-state testing Everything test related
#26166 opened Jul 27, 2026 by ggerganov Member Loading…
common: fix explicit -md precedence over draft sidecar resolution
#26165 opened Jul 27, 2026 by ServeurpersoCom Contributor Loading…
3
[Model/VLA] Support MiniCPM-RobotManip build Compilation issues documentation Improvements or additions to documentation mtmd Related to multimodal functionality (video/image/audio) server
#26164 opened Jul 27, 2026 by tc-mb Contributor • Draft
opencl: bugfix increment ref_count in ggml_backend_opencl_init() ggml changes relating to the ggml tensor library for machine learning OpenCL Issues specific to the OpenCL backend
#26162 opened Jul 27, 2026 by akleine Contributor Loading…
cuda: fuse MoE expert weighting and reduction (+3% to +9% prefill, MoE models only) CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning
#26160 opened Jul 27, 2026 by coder543 Contributor • Draft
1
cuda: compact Blackwell NVFP4 MoE work scheduling (+10% to +15% prefill, NVFP4-only) CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning
#26159 opened Jul 27, 2026 by coder543 Contributor • Draft
2
ggml: add support for MXFP8 CPU conversion ggml changes relating to the ggml tensor library for machine learning
#26157 opened Jul 27, 2026 by michaelw9999 Contributor Loading…
mtmd: support multi-row batching for deepseek-ocr mtmd Related to multimodal functionality (video/image/audio)
#26154 opened Jul 26, 2026 by ngxson Collaborator Loading…
2
skills: add ggml-test skill documentation Improvements or additions to documentation
#26153 opened Jul 26, 2026 by ngxson Collaborator Loading…
Make the WebGPU backend pass test-llama-archs tests ggml changes relating to the ggml tensor library for machine learning model Model specific testing Everything test related WebGPU
#26146 opened Jul 26, 2026 by fairydreaming Collaborator • Draft
Responses API: Convert text.format JSON schema into the corresponding Chat Completions response_format server testing Everything test related
#26145 opened Jul 26, 2026 by boondocklabs Contributor Loading…
1
ggml-cuda : fall back to cuBLAS when no MMQ tile size fits in shared memory CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning
#26141 opened Jul 26, 2026 by KakaruHayate Loading…
1
feat: QFX16/QFX32 lossless quantization (2.05x compression, zero quality loss) Apple Metal https://en.wikipedia.org/wiki/Metal_(API) examples ggml changes relating to the ggml tensor library for machine learning
#26136 opened Jul 26, 2026 by theQarchitect Loading…
2
ggml-webgpu: simplify flash_attn shaders and refactor several WGSL shaders ggml changes relating to the ggml tensor library for machine learning WebGPU
#26134 opened Jul 26, 2026 by yomaytk Member • Draft
ggml-cpu : fix conv_transpose_2d for multiple batches ggml changes relating to the ggml tensor library for machine learning testing Everything test related
#26132 opened Jul 26, 2026 by tekinertekin Contributor Loading…
1
server : expose per-device memory usage on /metrics and GET /memory documentation Improvements or additions to documentation mtmd Related to multimodal functionality (video/image/audio) server
#26130 opened Jul 26, 2026 by tobocop2 Loading…
1
2
vulkan: extend topk_moe fusion to support sqrt(softplus) ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#26124 opened Jul 25, 2026 by jeffbolznv Contributor Loading…
common: Add CLI > ENV > models-presets > INI precedence documentation Improvements or additions to documentation
#26118 opened Jul 25, 2026 by jcmdln Loading…
3
Fix speculative models failing to load when running llama-server
#26114 opened Jul 25, 2026 by sheldonrobinson Contributor Loading…
1
hexagon: enable quantized matmul for multi-sequence inputs ggml changes relating to the ggml tensor library for machine learning Hexagon
#26113 opened Jul 25, 2026 by w1049 Contributor Loading…
1
CUDA : add warp-per-row WKV7 kernel for single-token decode CUDA Related to the CUDA backend ggml changes relating to the ggml tensor library for machine learning testing Everything test related
#26111 opened Jul 25, 2026 by 123123213weqw Loading…
2
Feature/e2k support ggml changes relating to the ggml tensor library for machine learning
#26107 opened Jul 25, 2026 by png-tech Loading…
1
sycl: fix classification of iGPUs ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#26105 opened Jul 25, 2026 by KyleHagy Contributor Loading…
cmake : fall back to empty UI on stale assets when provisioning is disabled server/ui testing Everything test related
#26104 opened Jul 25, 2026 by wuisabel-gif • Draft
1 task done
1
3
ggml-cpu: skip unsupported ARM ISA variants under GGML_CPU_ALL_VARIANTS ggml changes relating to the ggml tensor library for machine learning
#26103 opened Jul 25, 2026 by ninihuang2026 Loading…
1
1
ProTip! Follow long discussions with comments:>50.