Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
863 Open 9,901 Closed
863 Open 9,901 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

(Performance; ggml-cpu) Optimized x86 and generic cpu q1_0 dot (follow up)
#21636 opened Apr 8, 2026 by pl752 • Draft
CUDA: also store node->src->data ptrs for equality check
#21635 opened Apr 8, 2026 by am17an Loading…
1
3
common : skip non-primary GGUF split files when selecting model
#21633 opened Apr 8, 2026 by angt Loading…
Enable ccache on riscv64
#21632 opened Apr 8, 2026 by luhenry • Draft
1
cmake: fix CMP0194 warning on Windows with MSVC
#21630 opened Apr 8, 2026 by texasich Loading…
1
cuda: Q1_0 initial backend
#21629 opened Apr 8, 2026 by khosravipasha Loading…
1
webui: Fix Model Selector choice sync
#21628 opened Apr 8, 2026 by allozaur Loading…
1
ci: fix labeler.yml v6 syntax — drop all: composition
#21627 opened Apr 8, 2026 by Marxist-Leninist Loading…
1
model: fix multimodal padding token for gemma3n/gemma4
#21625 opened Apr 8, 2026 by ngxson Loading…
feat: jinja engine improvements for reka-edge
#21623 opened Apr 8, 2026 by kwajiehao Loading…
4
SYCL: fix reorder crash when device memory is full
#21618 opened Apr 8, 2026 by PMZFX Loading…
5 tasks done
requirements : update transformers to 5.5.0
#21617 opened Apr 8, 2026 by danbev Loading…
2
Add support for Reka Edge 2603
#21616 opened Apr 8, 2026 by kwajiehao Loading…
12
gemma4: derive attn_soft_cap from GGUF instead of hardcoding
#21613 opened Apr 8, 2026 by stephencox-ict • Draft
4
CUDA: use a ring-buffer for cuda graphs
#21611 opened Apr 8, 2026 by am17an • Draft
vulkan: unify type macros to use Vx instead of _VECx ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#21605 opened Apr 8, 2026 by 0cc4m Loading…
1
quant: force Q6_K minimum for Gemma4 tied embeddings
#21599 opened Apr 8, 2026 by stephencox-ict • Draft
6 tasks done
3
SYCL: fix multi-GPU system RAM exhaustion by using Level Zero allocations ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#21597 opened Apr 8, 2026 by PMZFX Loading…
7 tasks done
26
opencl: add q5_K gemm and gemv kernels for Adreno ggml changes relating to the ggml tensor library for machine learning OpenCL Issues specific to the OpenCL backend
#21595 opened Apr 8, 2026 by shaofeiqi Loading…
common/sampling: reset reasoning budget sampler state between generations
#21594 opened Apr 8, 2026 by cnsiva Loading…
3
opencl: add basic support for q5_k ggml changes relating to the ggml tensor library for machine learning OpenCL Issues specific to the OpenCL backend
#21593 opened Apr 7, 2026 by shaofeiqi Loading…
vocab : fix Gemma 4 BPE tokenizer SIGSEGV on long prompts without newlines examples server testing Everything test related
#21587 opened Apr 7, 2026 by prue-starfield Loading…
2
fix: support non-ASCII (Unicode) file paths on Windows ggml changes relating to the ggml tensor library for machine learning
#21583 opened Apr 7, 2026 by lekot Loading…
3 tasks done
1
2
SYCL: add BF16 to DMMV kernel path (~4x tg speedup on Intel Arc) ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#21580 opened Apr 7, 2026 by PMZFX Loading…
1
webui: add "Send message on Enter" setting examples server
#21577 opened Apr 7, 2026 by mourix Loading…
2
ProTip! Exclude everything labeled bug with -label:bug.