ggml-org/llama.cpp

Pull Requests Commits

fit : add hp_ngl_skip to common_get_device_memory_data

am17an committed 3 days ago

08423882

fit : pass hp_ngl_skip by reference like the other hparams

am17an committed 3 days ago

4e2a188d

fit : account for skipped NextN layers when setting n_gpu_layers

am17an committed 4 days ago

159c23f2

dflash: copy hidden features via device copies (#31506)

am17an committed 4 days ago

Verified 2cbca389

docs: update s390x build docs (#31546)

taronaeo committed 4 days ago

Verified 29c720ea

mtmd: Pocket tts - support for voices without voice cloning (#31429)

rjtokenring committed 4 days ago

Verified d02b4c0d

model: add Clover support (#31306)

Maphist0 committed 4 days ago

Verified e4b6c8ca

llama : flash_attn_ext_rows - more efficient unified kv for multi-seq (#29510)

am17an committed 4 days ago

Verified 1b6935b2

imatrix : fix --nextn crash (#31522)

sanmai committed 4 days ago

Verified 3f8af15a

context : simplify setters (#29730)

nikwen committed 5 days ago

Verified cd8427d0

readme : fix links in README.md for maintainer PRs (#31487)

maternion committed 5 days ago

Verified 6b00d562

docs: CANN backend supports BF16 on 910B, document GGML_CANN_NO_PINNED (#31249)

li-lizhe committed 5 days ago

Verified 458fcfc3

ggml-zdnn: share buffers with host (#31481)

taronaeo committed 5 days ago

Verified 6ed81205

ggml : use `vm_allocate` instead of `posix_memalign` on all Apple platforms (#31437)

ykhrustalev committed 5 days ago

Verified ea24b952

opencl: add topk_mask (#31441)

lhez committed 5 days ago

Verified 84d9ce0f

vulkan: add tensor-parallel allreduce function (#25051)

pwilkin committed 5 days ago

Verified dcbb0916

vulkan: improve small-n split-K selection on RDNA3 and RDNA4 (#30146)

ravel7524 committed 5 days ago

Verified 07ec7c8a

vulkan: avoid redundant scans when zeroing skipped rows (#31266)

SG-Amadeus committed 5 days ago

Verified cb02250b

sycl : fast ggml_sycl_comm_exchange() on Intel Arc Pro B70s (#30956)

92es committed 5 days ago

Verified 2c93f0b6

sycl: dmmv code cleanup (#31325)

Titaniumtown committed 5 days ago

Verified 5e18eb02

ui: disable model deletion the server would refuse (#31305)

allozaur committed 5 days ago

Verified 5c2ea1ab

ui: align alert dialog styling with other overlay surfaces (#31312)

allozaur committed 5 days ago

Verified 23426d79

ui: keep queued steering messages in their own conversation (#31304)

allozaur committed 5 days ago

Verified 90080ce0

ui: show interrupted state for tool calls that never returned (#31314)

allozaur committed 5 days ago

Verified 64be95b1

ui: don't reload a loading model on select, prefer it in pre-selection (#31326)

allozaur committed 5 days ago

Verified b12b405c

docs : fix Vulkan example command in build.md (#31530)

Septa2112 committed 5 days ago

Verified aac6da80

tests : merge identical if branches (#31523)

nikwen committed 5 days ago

Verified 00ef1366

jinja : implement map('filter') (#24033)

MuoDoo committed 5 days ago

Verified 59695669

ggml : fix clang-cl build with AVX512 BF16 (#31276)

YagUber committed 5 days ago

Verified a7a490a4

docs: fix Huawei ascend brand (#31463)

licy666 committed 5 days ago

Verified f72dc309

Older