Search results

ggml : implement GEGLU_ERF and GEGLU_QUICK ops

Apple Metalhttps://en.wikipedia.org/wiki/Metal_(API)https://en.wikipedia.org/wiki/Metal_(API)

examples

ggmlchanges relating to the ggml tensor library for machine learningchanges relating to the ggml tensor library for machine learning

Nvidia GPUIssues specific to Nvidia GPUsIssues specific to Nvidia GPUs

SYCLhttps://en.wikipedia.org/wiki/SYCL - GPU programming languagehttps://en.wikipedia.org/wiki/SYCL - GPU programming language

VulkanIssues specific to the Vulkan backendIssues specific to the Vulkan backend

Status: Open (in progress).

ggml-org/llama.cppnumber 14445

#14445 In ggml-org/llama.cpp;

· CISC opened on Jun 29, 2025

webui: preserve partial content when streaming errors occur

examples

server

Status: Open (in progress).

ggml-org/llama.cppnumber 14374

#14374 In ggml-org/llama.cpp;

· Aaryan-549 opened on Jun 25, 2025

llama : add high-throughput mode

Apple Metalhttps://en.wikipedia.org/wiki/Metal_(API)https://en.wikipedia.org/wiki/Metal_(API)

examples

ggmlchanges relating to the ggml tensor library for machine learningchanges relating to the ggml tensor library for machine learning

Status: Draft (not ready).

ggml-org/llama.cppnumber 14363

#14363 In ggml-org/llama.cpp;

· ggerganov opened on Jun 24, 2025

server : fix assistant prefilling when content is an array

examples

pythonpython script changespython script changes

server

Status: Open (in progress).

ggml-org/llama.cppnumber 14360

#14360 In ggml-org/llama.cpp;

· CISC opened on Jun 24, 2025

[server] webui DB import and export

examples

server

Status: Open (in progress).

ggml-org/llama.cppnumber 14347

#14347 In ggml-org/llama.cpp;

· feiloo opened on Jun 23, 2025

make "server-core" library

examples

server

Status: Open (in progress).

ggml-org/llama.cppnumber 14331

#14331 In ggml-org/llama.cpp;

· arnej27959 opened on Jun 22, 2025

GitHub workflow: set RPATH to "@loader_path" / "$ORIGIN" to ensure executables and dynamic libraries search for dependencies in their origin directory.

devopsimprovements to build systems and github actionsimprovements to build systems and github actions

examples

Status: Open (in progress).

ggml-org/llama.cppnumber 14309

#14309 In ggml-org/llama.cpp;

· rotemdan opened on Jun 20, 2025

kv-cache : use ggml_set_rows

Apple Metalhttps://en.wikipedia.org/wiki/Metal_(API)https://en.wikipedia.org/wiki/Metal_(API)

examples

ggmlchanges relating to the ggml tensor library for machine learningchanges relating to the ggml tensor library for machine learning

testingEverything test relatedEverything test related

Status: Draft (not ready).

ggml-org/llama.cppnumber 14285

#14285 In ggml-org/llama.cpp;

· ggerganov opened on Jun 19, 2025

server : add pidfile option

examples

server

Status: Open (in progress).

ggml-org/llama.cppnumber 14242

#14242 In ggml-org/llama.cpp;

· ericcurtin opened on Jun 17, 2025

Mtmd: add a way to select device for vision encoder

examples

Status: Open (in progress).

ggml-org/llama.cppnumber 14236

#14236 In ggml-org/llama.cpp;

· stduhpf opened on Jun 17, 2025

ggml: introduce GGML_NUMA_MIGRATE to optimize cross NUMA op computation

examples

ggmlchanges relating to the ggml tensor library for machine learningchanges relating to the ggml tensor library for machine learning

Status: Open (in progress).

ggml-org/llama.cppnumber 14232

#14232 In ggml-org/llama.cpp;

· wenlujon opened on Jun 17, 2025

logit_bias: apply configurable escalating EOG bias at low n_remain

examples

server

testingEverything test relatedEverything test related

Status: Open (in progress).

ggml-org/llama.cppnumber 14229

#14229 In ggml-org/llama.cpp;

· graehl opened on Jun 16, 2025