Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
1,039 Open 10,849 Closed
1,039 Open 10,849 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

hexagon: add support for CONCAT op
#23648 opened May 25, 2026 by max-krasnyansky Member • Draft
hexagon: add support for Q4_1 in MUL_MAT and MUL_MAT_ID ggml changes relating to the ggml tensor library for machine learning Hexagon
#23647 opened May 25, 2026 by max-krasnyansky Member • Draft
server: MTP layer kv-cache should respect draft type ctk examples merge ready A maintainer can use this label to indicate that they consider the changes final and ready to merge. server
#23646 opened May 25, 2026 by am17an Contributor Loading…
2
fix: sanitize sampling and mirostat parameters to prevent unstable states
#23644 opened May 25, 2026 by gustavo89587 Loading…
1
llama: add llm_graph_input_mtp model Model specific
#23643 opened May 25, 2026 by am17an Contributor • Draft
1
ci: update spacemit toolchain url and enhance curl command devops improvements to build systems and github actions documentation Improvements or additions to documentation
#23642 opened May 25, 2026 by alex-spacemit Collaborator Loading…
4
vulkan: don't hold the device mutex while compiling pipelines ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#23641 opened May 25, 2026 by jeffbolznv Contributor Loading…
Server: avoid evicting active/loading models in router LRU examples server
#23640 opened May 25, 2026 by loopy321 Loading…
vendor : update cpp-httplib to 0.45.1 python python script changes script Script related
#23639 opened May 25, 2026 by cabelo Contributor Loading…
mtmd : enable dynamic hi-res tiling for nemotron_v2_vl (Nemotron Nano Omni) examples
#23638 opened May 25, 2026 by SyrupAnon Loading…
1 task done
1
tests: test-backend-ops -j <N> to run tests in parallel testing Everything test related
#23637 opened May 25, 2026 by jeffbolznv Contributor Loading…
ci : install host compiler on android-ndk build devops improvements to build systems and github actions
#23630 opened May 24, 2026 by aldehir Contributor Loading…
Fix 23627: Attach Mistral3 NVFP4 weight scales model Model specific
#23629 opened May 24, 2026 by michaelw9999 Contributor Loading…
gguf-py: preserve MoE size labels for mmproj metadata python python script changes
#23618 opened May 24, 2026 by ooovenenoso Loading…
1
CUDA: add fast walsh-hadamard transform ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs testing Everything test related
#23615 opened May 24, 2026 by am17an Contributor Loading…
cuda : fix KQ mask offset integer overflow in flash attention MMA kernel ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#23610 opened May 24, 2026 by fairydreaming Collaborator Loading…
1
1
cuda: read memory through NVML if available ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#23604 opened May 24, 2026 by 0cc4m Contributor Loading…
11
common: preserve horizontal whitespace in tool calls testing Everything test related
#23602 opened May 24, 2026 by Krish2882005 Loading…
1 task done
1
webui: added single line reasoning preview examples server/ui
#23601 opened May 24, 2026 by gugugiyu Loading…
1
Parallelize quant LUT init ggml changes relating to the ggml tensor library for machine learning
#23595 opened May 24, 2026 by jeffbolznv Contributor Loading…
@ggerganov
ggml-webgpu: Add MMVQ path for Q4/Q8/Q2_K/Q4_K and clean up legacy MUL_MAT pipeline ggml changes relating to the ggml tensor library for machine learning WebGPU
#23594 opened May 24, 2026 by yomaytk Contributor Loading…
12
ggml: fix AVX-512 BF16 build with clang-cl ggml changes relating to the ggml tensor library for machine learning
#23593 opened May 24, 2026 by marcusds Loading…
1
Update build.md with Fedora Vulkan dependencies documentation Improvements or additions to documentation
#23584 opened May 23, 2026 by JCTRoth Loading…
cmake : error when LLAMA_BUILD_APP=ON and LLAMA_BUILD_TOOLS=OFF build Compilation issues
#23580 opened May 23, 2026 by Pento95 Loading…
Static quantize mtp layers
#23575 opened May 23, 2026 by de-wim • Draft
3
ProTip! Add no:assignee to see everything that’s not assigned.