ngxson/llama.cpp

Pull Requests Commits

tests: Avoid floating point precision false positives in SUM (#17471)

jeffbolznv committed 7 days ago

Verified 52ab19df

test-backend-ops: improve msvc build time (#18209)

jeffbolznv committed 7 days ago

Verified 5182dd64

Added comments explaining thread block size selection logic based on row count and column size, derived from historical commit context (#18212)

Aadeshveer committed 8 days ago

Verified 10b4f82d

server : [easy] fix per round speculative decode logging (#18211)

okuvshynov committed 8 days ago

Verified 408616ad

server: support load model on startup, support preset-only options (#18206)

ngxson committed 8 days ago

Verified 9e39a1e6

ci : remove non-windows zip artifacts (#18201)

CISC committed 8 days ago

Verified 74e05131

ci : only save ccache on master (#18207)

CISC committed 8 days ago

Verified f74747d8

ggml-hexagon: Implement true Q8_0 quantization on Hexagon NPU for more accurate mixed-precision matmul operations (#17977)

ngdxzy committed 8 days ago

Verified ce734a8a

arg: fix order to use short form before long form (#18196)

ServeurpersoCom committed 8 days ago

Verified 14931a82

llama : Changing off_t to size_t for Windows (#18204)

JTischbein committed 9 days ago

Verified f99ef53d

server: friendlier error msg when ctx < input (#18174)

am17an committed 9 days ago

Verified cc0a0434

presets: refactor, allow cascade presets from different sources, add global section (#18169)

ngxson committed 9 days ago

Verified 98c1c7a7

webui: Add editing attachments in user messages (#18147)

allozaur committed 9 days ago

Verified acb73d83

model-conversion : add verbose flag in run-org-model.py (#18194)

danbev committed 9 days ago

Verified 0a271d82

android: fix missing screenshots for Android.md (#18156)

naco-siren committed 9 days ago

Verified 52fc7fee

vulkan: Add perf logger mode with concurrency (#17944)

jeffbolznv committed 9 days ago

Verified cdbada8d

model : add ASR support for LFM2-Audio-1.5B (conformer) (#18106)

ngxson committed 9 days ago

Verified 8ea958d4

webui: display prompt processing stats (#18146)

ServeurpersoCom committed 9 days ago

Verified f9ec8858

ggml-cpu: extend support for RVV floating-point kernels (#17318)

taimur-10x committed 10 days ago

Verified f716588e

arg: fix ASAN error on sampler_type_names empty (#18167)

ngxson committed 10 days ago

Verified 4d1316c4

gguf-py : use copy-on-write mode for localtensor (#18162)

CISC committed 10 days ago

Verified ec7b9329

remove i_major_dual (#18157)

zhang-hui-yulo committed 10 days ago

Verified 54189c0d

webui: Fix selecting generated output issues during active streaming (#18091)

allozaur committed 10 days ago

Verified 9ce64aed

webui: fix chat screen shadow width (#18010)

polydecay committed 10 days ago

Verified 900316da

llama: offload output layer to GPU first (#18148)

JohannesGaessler committed 10 days ago

Verified 57c1e056

convert : sort and use file parts from model index if present (#18043)

CISC committed 10 days ago

Verified 9cff4cc5

llama : Async DirectIO model loading on Linux (#18012)

JTischbein committed 10 days ago

Verified 4d4f4cac

ggml-hexagon: swiglu_oai operation (#18114)

joeldushouyu committed 10 days ago

Verified 0a0bba05

convert : force patch_merger tensors to f16/f32 (#18124)

CISC committed 10 days ago

Verified 5166aaf8

server: (webui) add --webui-config (#18028)

ServeurpersoCom committed 10 days ago

Verified 6ce3d857

Newer Older