ngxson/llama.cpp

Pull Requests Commits

Apply suggestions from code review

ngxson committed 149 days ago

Verified 8c543f7c

correct project activation fn

ngxson committed 149 days ago

4556b403

minor coding style improvement

ngxson committed 149 days ago

01bf6872

fix regression for ultravox

ngxson committed 149 days ago

8b2d72da

add docs and tests

ngxson committed 149 days ago

738be198

also support Devstral conversion

ngxson committed 149 days ago

b828887a

add [BEGIN_AUDIO] token

ngxson committed 149 days ago

97119dd7

Merge branch 'master' into xsn/voxtral

ngxson committed 149 days ago

49045bd3

quantize : update README.md (#14905)

EAddario committed 149 days ago

Verified 7f975995

vulkan: add ops docs (#14900)

0cc4m committed 150 days ago

Verified bf78f543

SYCL: add ops doc (#14901)

qnixsynapse committed 150 days ago

Verified bbfc8492

llama : clarify comment about pp and tg graphs [no ci] (#14895)

danbev committed 150 days ago

Verified ca0ef2dd

vulkan : add fp16 support for the conv_2d kernel (#14872)

Green-Sky committed 150 days ago

Verified 89d10295

vulkan: skip empty set_rows to avoid invalid API usage (#14860)

jeffbolznv committed 150 days ago

Verified f1a4e72d

model : make rope_yarn_log_mul optional for deepseek2 (#14896)

gabriellarson committed 150 days ago

Verified 4762ad73

llama : fix kq_scale for the attention layers of PLaMo2 (#14892)

mitmul committed 150 days ago

Verified 1dc9614e

Docs: add instructions for adding backends (#14889)

am17an committed 150 days ago

Verified 446595b9

HIP: Enable Matrix cores for MMQ Kernels, Enable stream-K for CDNA 3 (#14624)

deepsek committed 150 days ago

Verified 66906cd8

CANN: Implement GLU ops (#14884)

hipudding committed 151 days ago

Verified 11dd5a44

musa: fix build warnings (unused variable) (#14869)

yeahdongcn committed 151 days ago

Verified 9b8f3c6c

ggml-cpu : disable GGML_NNPA by default due to instability (#14880)

taronaeo committed 152 days ago

Verified c7f3169c

metal: SSM_SCAN performance (#14743)

gabe-l-hart committed 152 days ago

Verified 793c0d7f

opencl: add fused `rms_norm_mul` (#14841)

lhez committed 152 days ago

Verified ce111d39

docs : update HOWTO‑add‑model.md for ModelBase and new model classes (#14874)

wooksong committed 152 days ago

Verified e7fecba9

ggml : remove invalid portPos specifiers from dot files (#14838)

ORippler committed 152 days ago

Verified e2b7621e

context : restore preemptive sched reset when LLAMA_SET_ROWS=0 (#14870)

ggerganov committed 152 days ago

Verified c1dbea75

mtmd : fix 32-bit narrowing issue in export-lora and mtmd clip (#14503)

kiwi142857 committed 152 days ago

Verified 749e0d27

rpc : check for null buffers in get/set/copy tensor endpoints (#14868)

struct committed 152 days ago

Verified 64bf1c37

sched : fix multiple evaluations of the same graph with pipeline parallelism (#14855)

slaren committed 152 days ago

Verified c12bbde3

fix python requirements

ngxson committed 152 days ago

2da31edd

Older