-
Notifications
You must be signed in to change notification settings - Fork 14.1k
Pull requests: ggml-org/llama.cpp
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
server: validate n_batch == n_ubatch for embeddings (#6263)
#18123
opened Dec 17, 2025 by
yifant-code
•
Draft
[WIP] Cross Entropy Loss on Metal
Apple Metal
https://en.wikipedia.org/wiki/Metal_(API)
ggml
changes relating to the ggml tensor library for machine learning
ci : clean up webui jobs
devops
improvements to build systems and github actions
#18116
opened Dec 16, 2025 by
CISC
Loading…
ggml-hexagon: swiglu_oai operation
ggml
changes relating to the ggml tensor library for machine learning
#18114
opened Dec 16, 2025 by
joeldushouyu
Loading…
ggml-hexagon: Add lightweight atomic synchronization support to htp_ops_context for inter-task coordination
ggml
changes relating to the ggml tensor library for machine learning
#18113
opened Dec 16, 2025 by
ngdxzy
Loading…
server: (router) allow child process to report status via stdout
examples
server
#18110
opened Dec 16, 2025 by
ngxson
Loading…
model : add ASR support for LFM2-Audio-1.5B (conformer)
examples
ggml
changes relating to the ggml tensor library for machine learning
Nvidia GPU
Issues specific to Nvidia GPUs
python
python script changes
testing
Everything test related
#18106
opened Dec 16, 2025 by
ngxson
Loading…
ggml-cuda: Delta-Net linear attention for Qwen3-Next
ggml
changes relating to the ggml tensor library for machine learning
model
Model specific
Nvidia GPU
Issues specific to Nvidia GPUs
testing
Everything test related
#18102
opened Dec 16, 2025 by
hauhaut
Loading…
ggml-cpu: ARM64: repack version of q8_0 (dotprod and i8mm)
ggml
changes relating to the ggml tensor library for machine learning
#18096
opened Dec 16, 2025 by
Alcpz
Loading…
server: [RFC] add optional POST /exit endpoint for graceful shutdown
examples
server
#18086
opened Dec 16, 2025 by
qnixsynapse
•
Draft
ggml: migrate work_data to stack allocation
ggml
changes relating to the ggml tensor library for machine learning
testing
Everything test related
#18083
opened Dec 16, 2025 by
GermanAizek
Loading…
vulkan/cuda: fix topk_moe with exp_probs_b
ggml
changes relating to the ggml tensor library for machine learning
Nvidia GPU
Issues specific to Nvidia GPUs
testing
Everything test related
Vulkan
Issues specific to the Vulkan backend
#18071
opened Dec 15, 2025 by
jeffbolznv
Loading…
webui: add responsive chat width option to webui (#18067)
examples
server
#18068
opened Dec 15, 2025 by
ImadSaddik
Loading…
vulkan: support GGML_UNARY_OP_XIELU
ggml
changes relating to the ggml tensor library for machine learning
Vulkan
Issues specific to the Vulkan backend
#18062
opened Dec 15, 2025 by
jeffbolznv
Loading…
vulkan: in graph_optimize, try to group ADD operations
ggml
changes relating to the ggml tensor library for machine learning
Vulkan
Issues specific to the Vulkan backend
#18060
opened Dec 15, 2025 by
jeffbolznv
Loading…
webui: Client-side implementation of tool calling (with two tools)
examples
server
#18059
opened Dec 15, 2025 by
coder543
Loading…
common: fix --override-kv to support comma-separated values
#18056
opened Dec 15, 2025 by
ServeurpersoCom
Loading…
CLI: llama-cli and llama-completion cosmetics
devops
improvements to build systems and github actions
documentation
Improvements or additions to documentation
python
python script changes
script
Script related
SYCL
https://en.wikipedia.org/wiki/SYCL - GPU programming language
#18053
opened Dec 15, 2025 by
andrew-aladev
Loading…
vulkan: Implement set_tensor_async and the event interfaces
ggml
changes relating to the ggml tensor library for machine learning
Vulkan
Issues specific to the Vulkan backend
#18047
opened Dec 15, 2025 by
jeffbolznv
Loading…
convert : keep file part order from model index
python
python script changes
#18043
opened Dec 14, 2025 by
CISC
Loading…
[Speculative decoding] feat: add EAGLE3 speculative decoding support
examples
ggml
changes relating to the ggml tensor library for machine learning
model
Model specific
python
python script changes
#18039
opened Dec 14, 2025 by
ichbinhandsome
•
Draft
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.