Search results

vulkan: Split large mul_mat_id to fit in shared memory

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

testingEverything test relatedEverything test related

VulkanIssues specific to the Vulkan backendIssues specific to the Vulkan backend

Status: Open (in progress).

ggml-org/llama.cppnumber 14451

#14451 In ggml-org/llama.cpp;

· jeffbolznv opened on Jun 29, 2025

ggml : support broadcast for ggml_soft_max_ext and ggml_flash_attn_ext

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

Ascend NPUissues specific to Ascend NPUsissues specific to Ascend NPUs

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

testingEverything test relatedEverything test related

VulkanIssues specific to the Vulkan backendIssues specific to the Vulkan backend

Status: Open (in progress).

ggml-org/llama.cppnumber 14435

#14435 In ggml-org/llama.cpp;

· ggerganov opened on Jun 28, 2025

ggml : add ggml_scale_bias

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

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 14417

#14417 In ggml-org/llama.cpp;

· ngxson opened on Jun 27, 2025

test-backend-ops: add support for specifying output format

testingEverything test relatedEverything test related

Status: Open (in progress).

ggml-org/llama.cppnumber 14368

#14368 In ggml-org/llama.cpp;

· yeahdongcn opened on Jun 25, 2025

ggml: adds CONV_2D op and direct GEMM Vulkan implementation

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

testingEverything test relatedEverything test related

VulkanIssues specific to the Vulkan backendIssues specific to the Vulkan backend

Status: Open (in progress).

ggml-org/llama.cppnumber 14316

#14316 In ggml-org/llama.cpp;

· etasnadi opened on Jun 21, 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

MODEL: Falcon-H1 support

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

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

pythonpython script changespython script changes

testingEverything test relatedEverything test related

Status: Draft (not ready).

ggml-org/llama.cppnumber 14238

#14238 In ggml-org/llama.cpp;

· younesbelkada 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

tests : add test-model-random

help wantedExtra attention is neededExtra attention is needed

testingEverything test relatedEverything test related

Status: Draft (not ready).

ggml-org/llama.cppnumber 14139

#14139 In ggml-org/llama.cpp;

· compilade opened on Jun 12, 2025

<code>chat</code>: improve llama 3.x handling of &lt;|python_tag|&gt; (+ allow --special combo)

testingEverything test relatedEverything test related

Status: Draft (not ready).

ggml-org/llama.cppnumber 13932

#13932 In ggml-org/llama.cpp;

· ochafik opened on May 30, 2025

finetune.cpp command-line arg

buildCompilation issuesCompilation issues

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

testingEverything test relatedEverything test related

Status: Open (in progress).

ggml-org/llama.cppnumber 13873

#13873 In ggml-org/llama.cpp;

· graehl opened on May 28, 2025

ggml : add ggml_fill()

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: Open (in progress).

ggml-org/llama.cppnumber 13772

#13772 In ggml-org/llama.cpp;

· ngxson opened on May 25, 2025