ngxson/llama.cpp

Pull Requests Commits

linenoise.cpp refactoring (#11301)

ericcurtin committed 360 days ago

Verified 2e2f8f09

metal : fix out-of-bounds write (#11314)

ggerganov committed 360 days ago

Verified 2139667e

common : add -hfd option for the draft model (#11318)

ggerganov committed 360 days ago

Verified 80d0d6b4

vulkan: fix coopmat2 validation failures (#11284)

jeffbolznv committed 360 days ago

Verified aea8ddd5

examples : fix add_special conditions (#11311)

ggerganov committed 360 days ago

Verified 9f7add1c

mmap: add include for cerrno (#11296)

mascguy committed 360 days ago

Verified 90d987b1

cmake: fix shell command quoting in build-info script (#11309)

Xarbirus committed 360 days ago

Verified a4251edd

llama : add support for Deepseek-R1-Qwen distill model (#11310)

ngxson committed 360 days ago

Verified ec7f3ac9

cont : fix whitespaces (#11305)

ggerganov committed 361 days ago

Verified ef6dada6

llama : re-add LLM_ARCH_PHIMOE (#11305)

KyleBruene committed 361 days ago

Verified ae3c1db2

tests : increase timeout when sanitizers are enabled (#11300)

ggerganov committed 361 days ago

Verified 92bc4939

simple-chat : fix BOS being added to each message (#11278)

ggerganov committed 361 days ago

Verified b9daaffe

SYCL: Introducing memory host pool (#11251)

s-Nick committed 361 days ago

Verified 99487b57

Adding linenoise.cpp to llama-run (#11252)

ericcurtin committed 362 days ago

Verified a1649cc1

cmake : add sanitizer flags for llama.cpp (#11279)

ggerganov committed 362 days ago

Verified 4dd34ff8

server : implement cancellable request (#11285)

ngxson committed 362 days ago

Verified f30f0992

scripts : restore hf.sh (#11288)

ggerganov committed 363 days ago

Verified f26c8741

tts : add guide tokens support (#11186)

LostRuins committed 363 days ago

Verified 6390a998

vulkan: fix coopmat2 flash attention for non-contiguous inputs (#11281)

jeffbolznv committed 363 days ago

Verified 44e18ef9

llama.android: add field formatChat to control whether to parse special tokens when send message (#11270)

codezjx committed 363 days ago

Verified 3edfa7d3

rpc : early register backend devices (#11262)

rgerganov committed 364 days ago

Verified 667d7284

vocab : fix double-eos check (#11273)

ggerganov committed 364 days ago

Verified a133566d

llama : fix deprecation message: vocabable -> vocab (#11269)

dwrensha committed 364 days ago

Verified 960ec652

README : added kalavai to infrastructure list (#11216)

musoles committed 364 days ago

Verified 7a689c41

vulkan: support copy from f32 to q4_0/q4_1/q5_0/q5_1/q8_0/iq4_nl (#11166)

jeffbolznv committed 364 days ago

Verified bd38ddea

vulkan: optimize coopmat2 q4_k/q5_k dequant functions. (#11206)

jeffbolznv committed 364 days ago

Verified 466300fe

vulkan: optimize coopmat2 q2_k dequant function (#11130)

jeffbolznv committed 364 days ago

Verified 206bc534

llama : add internlm3 support (#11233)

RunningLeon committed 364 days ago

Verified 4dbc8b9c

CUDA: backwards pass for misc. ops, add tests (#11257)

JohannesGaessler committed 364 days ago

Verified 9c8dcefe

llama : add `llama_model_load_from_splits` (#11255)

ngxson committed 364 days ago

Verified 681149ce

Newer Older