Skip to content

Pull requests: ggml-org/llama.cpp

New pull request New
723 Open 8,984 Closed
723 Open 8,984 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

Q5_K - Block Interleaving Implementation for x86 SIMD (AVX512/AVX2) ggml changes relating to the ggml tensor library for machine learning
#19707 opened Feb 18, 2026 by Manogna-Sree Loading…
Q6_K - Block Interleaving Implementation for x86 SIMD (AVX512/AVX2) ggml changes relating to the ggml tensor library for machine learning
#19706 opened Feb 18, 2026 by Manogna-Sree Loading…
common : fix gpt-oss Jinja error with content and thinking on tool-call messages
#19704 opened Feb 18, 2026 by abhijitb11 Loading…
1
ggml-webgpu: Add unary op (SQR, SQRT, SIN, COS) support. documentation Improvements or additions to documentation ggml changes relating to the ggml tensor library for machine learning
#19700 opened Feb 18, 2026 by yomaytk Loading…
Add Mistral Voxtral Mini 4B Realtime 2602 4B streaming ASR support examples model Model specific python python script changes
#19698 opened Feb 17, 2026 by Acceldium Loading…
New option GGML_CUDA_FORCE_CUBLAS_COMPUTE_32F to use fp32 as compute type in cuBLAS documentation Improvements or additions to documentation ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#19697 opened Feb 17, 2026 by wallentri88 Loading…
server : fix V-L embedding model support
#19694 opened Feb 17, 2026 by oliveagle Loading…
test(server): add multi-image and no-image vision API tests examples python python script changes server
#19691 opened Feb 17, 2026 by jorgeutd Loading…
3 tasks done
Pre-MCP UI and architecture cleanup examples server
#19689 opened Feb 17, 2026 by allozaur Loading…
1
model : Add tokenizer from LFM2.5-Audio-1.5B model Model specific python python script changes
#19687 opened Feb 17, 2026 by tdakhran Loading…
16
CUDA: fix kernel selection logic for tile FA ggml changes relating to the ggml tensor library for machine learning Nvidia GPU Issues specific to Nvidia GPUs
#19686 opened Feb 17, 2026 by JohannesGaessler Loading…
1
1
DOCS: Fix broken links for preparing models in Backends Ascend NPU issues specific to Ascend NPUs documentation Improvements or additions to documentation SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language
#19684 opened Feb 17, 2026 by MaciejDromin Loading…
model: support GLM-OCR examples model Model specific python python script changes
#19677 opened Feb 16, 2026 by ngxson Loading…
1
4
Add Pylint workflow for Python code analysis devops improvements to build systems and github actions
#19671 opened Feb 16, 2026 by kerrrang9214-tech • Draft
Allow partial success of seq_rm for hybrid memory
#19670 opened Feb 16, 2026 by Nekotekina Loading…
Add Kimi Linear to unified delta net model Model specific
#19668 opened Feb 16, 2026 by ymcki Loading…
23
llama : use output_resolve_row() in get_logits_ith/get_embeddings_ith
#19663 opened Feb 16, 2026 by danbev Loading…
models : dedup qwen35 graphs model Model specific
#19660 opened Feb 16, 2026 by ggerganov • Draft
2 tasks
3
avx2: compute ksigns instead of loading from table ggml changes relating to the ggml tensor library for machine learning
#19657 opened Feb 16, 2026 by dfriehs Loading…
1
common : fix Step-3.5-Flash format detection and thinking support testing Everything test related
#19635 opened Feb 15, 2026 by jesseposner Loading…
8
Vulkan Scalar Flash Attention Refactor ggml changes relating to the ggml tensor library for machine learning Vulkan Issues specific to the Vulkan backend
#19625 opened Feb 14, 2026 by 0cc4m • Draft
6
[server] save generated text for the /slots endpoint (for LLAMA_SERVER_SLOTS_DEBUG=1) examples server
#19622 opened Feb 14, 2026 by matteoserva Loading…
2
Fix gpt-oss tool calling: pass tool args and tool responses as json
#19620 opened Feb 14, 2026 by matteoserva Loading…
1
3
quantize : refactor llama-quant.cpp examples
#19616 opened Feb 14, 2026 by ddh0 • Draft
fix: GLM 4.5 streaming tool-call parsing + grammar error handling examples server testing Everything test related
#19612 opened Feb 14, 2026 by Gunther-Schulz Loading…
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.