- Notifications
You must be signed in to change notification settings - Fork13.9k
Pull requests: ggml-org/llama.cpp
Author
Uh oh!
There was an error while loading.Please reload this page.
Label
Uh oh!
There was an error while loading.Please reload this page.
Projects
Uh oh!
There was an error while loading.Please reload this page.
Milestones
Uh oh!
There was an error while loading.Please reload this page.
Reviews
Assignee
Assigned to nobodyLoading
Uh oh!
There was an error while loading.Please reload this page.
Sort
Pull requests list
Document how to compile with Vulkan using Debian/Ubuntu packages
#17688 openedDec 2, 2025 bysocram8888Loading…
vulkan : support conv-2d with large output size ggmlchanges relating to the ggml tensor library for machine learning testingEverything test related VulkanIssues specific to the Vulkan backend
#17685 openedDec 2, 2025 byAclyLoading…
FIX:Adding the utf8 compilation option can fix this debug jump issue. ggmlchanges relating to the ggml tensor library for machine learning
#17682 openedDec 2, 2025 byxiaobing318Loading…
vulkan: enable mmvq for q2_k on NVIDIA ggmlchanges relating to the ggml tensor library for machine learning VulkanIssues specific to the Vulkan backend
#17675 openedDec 2, 2025 byjeffbolznvLoading…
vulkan: perf_logger improvements ggmlchanges relating to the ggml tensor library for machine learning VulkanIssues specific to the Vulkan backend
#17672 openedDec 2, 2025 byjeffbolznvLoading…
vulkan: fix top_k bug when there are ties in the input ggmlchanges relating to the ggml tensor library for machine learning testingEverything test related VulkanIssues specific to the Vulkan backend
#17659 openedDec 1, 2025 byjeffbolznvLoading…
ggml: use 'exists( const std::filesystem::path&, std::error_code&)' instead of 'exists( const std::filesystem::path&)' to enhance robustness ggmlchanges relating to the ggml tensor library for machine learning
#17653 openedDec 1, 2025 byflyinskyin2013Loading…
ggml: added missing cast sections in memcpy ggmlchanges relating to the ggml tensor library for machine learning vibe-codedCreated with heavy use of LLM assistants, requires human verification
#17651 openedDec 1, 2025 byGermanAizekLoading…
ggml-cpu: remove duplicate conditional check 'iid' ggmlchanges relating to the ggml tensor library for machine learning
#17650 openedDec 1, 2025 byGermanAizekLoading…
gguf: llama: usechanges relating to the ggml tensor library for machine learning vibe-codedCreated with heavy use of LLM assistants, requires human verification
= default for trivial constructors and destructors ggml #17649 openedDec 1, 2025 byGermanAizekLoading…
sgemm: reuse loaded vector in AVX dot product calculation ggmlchanges relating to the ggml tensor library for machine learning vibe-codedCreated with heavy use of LLM assistants, requires human verification
#17648 openedDec 1, 2025 byGermanAizekLoading…
llama-vocab: replace postfix with prefix increment for iterators vibe-codedCreated with heavy use of LLM assistants, requires human verification
#17646 openedDec 1, 2025 byGermanAizekLoading…
vec: optimize AVX2/FMA sum-of-squares with loop unrolling and FMA ggmlchanges relating to the ggml tensor library for machine learning vibe-codedCreated with heavy use of LLM assistants, requires human verification
#17642 openedDec 1, 2025 byGermanAizekLoading…
ggml-quants: use _mm256_testz_si256 for mask checks in AVX2 ggmlchanges relating to the ggml tensor library for machine learning vibe-codedCreated with heavy use of LLM assistants, requires human verification
#17641 openedDec 1, 2025 byGermanAizekLoading…
ggml-alloc: optimize free block shifting withchanges relating to the ggml tensor library for machine learning vibe-codedCreated with heavy use of LLM assistants, requires human verification
memmove ggml #17640 openedDec 1, 2025 byGermanAizekLoading…
vulkan: Replace deprecated VK_EXT_validation_features ggmlchanges relating to the ggml tensor library for machine learning VulkanIssues specific to the Vulkan backend
#17637 openedDec 1, 2025 byrillomasLoading…
common : compute average token length from vocabulary
#17632 openedDec 1, 2025 byyifant-code • Draft
llama-router, the C++ "llama-swap" for llama.cpp examples need feedbackTesting and feedback with results are needed server testingEverything test related
#17629 openedNov 30, 2025 byServeurpersoCom • Draft
vulkan: set all memory allocations to high priority ggmlchanges relating to the ggml tensor library for machine learning VulkanIssues specific to the Vulkan backend
#17624 openedNov 30, 2025 byjeffbolznv • Draft
vulkan: Reduce temporary memory usage for TOP_K ggmlchanges relating to the ggml tensor library for machine learning VulkanIssues specific to the Vulkan backend
#17623 openedNov 30, 2025 byjeffbolznvLoading…
ProTip! Mix and match filters to narrow down what you’re looking for.