ochafik/llama.cpp

Pull Requests Commits

llama : simplify mmap session cache with fixed-size files

claude committed 239 days ago

Verified 197ee652

docs: add session-mmap.md documentation

claude committed 247 days ago

Verified ba7acdb3

llama : add copy-on-write mode for mmap session cache

claude committed 247 days ago

Verified 75f92859

llama : add memory-mapped session cache support

claude committed 247 days ago

Verified 20d05e19

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

JTischbein committed 252 days ago

Verified f99ef53d

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

am17an committed 253 days ago

Verified cc0a0434

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

ngxson committed 253 days ago

Verified 98c1c7a7

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

allozaur committed 253 days ago

Verified acb73d83

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

danbev committed 253 days ago

Verified 0a271d82

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

naco-siren committed 253 days ago

Verified 52fc7fee

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

jeffbolznv committed 253 days ago

Verified cdbada8d

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

ngxson committed 253 days ago

Verified 8ea958d4

webui: display prompt processing stats (#18146)

ServeurpersoCom committed 253 days ago

Verified f9ec8858

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

taimur-10x committed 254 days ago

Verified f716588e

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

ngxson committed 254 days ago

Verified 4d1316c4

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

CISC committed 254 days ago

Verified ec7b9329

remove i_major_dual (#18157)

zhanghuiyulo committed 254 days ago

Verified 54189c0d

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

allozaur committed 254 days ago

Verified 9ce64aed

webui: fix chat screen shadow width (#18010)

polydecay committed 254 days ago

Verified 900316da

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

JohannesGaessler committed 254 days ago

Verified 57c1e056

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

CISC committed 254 days ago

Verified 9cff4cc5

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

JTischbein committed 254 days ago

Verified 4d4f4cac

ggml-hexagon: swiglu_oai operation (#18114)

joeldushouyu committed 254 days ago

Verified 0a0bba05

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

CISC committed 254 days ago

Verified 5166aaf8

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

ServeurpersoCom committed 254 days ago

Verified 6ce3d857

server: (router) disable SSL on child process (#18141)

ngxson committed 254 days ago

Verified e85e9d76

llama-fit-params: fix memory print (#18136)

JohannesGaessler committed 254 days ago

Verified 8dcc3662

webui: fix chat header width when sidebar is closed (#17981)

polydecay committed 254 days ago

Verified d37fc935

ggml-hexagon: gelu operation (#17921)

joeldushouyu committed 254 days ago

Verified 4470a076

common : restore grammar-based rejection sampling (#18137)

ggerganov committed 254 days ago

Verified 4301e273

Older