ggerganov/llama.cpp

Pull Requests Commits

ggml : fix possible buffer use after free in sched reserve

slaren committed 357 days ago

630bce5a

readme : update bindings list (#9918)

ShenghaiWang committed 358 days ago

Verified 3752217e

vulkan : add backend registry / device interfaces (#9721)

slaren committed 358 days ago

Verified f010b77a

fix: allocating CPU buffer with size `0` (#9917)

giladgd committed 358 days ago

Verified 21942002

fix: use `vm_allocate` to allocate CPU backend buffer on macOS (#9875)

giladgd committed 358 days ago

Verified 73afe681

llama : suppress conversion from 'size_t' to 'int' (#9046)

danbev committed 358 days ago

Verified 9e041024

llava : fix typo in error message [no ci] (#9884)

danbev committed 358 days ago

Verified dbf18e4d

grammar : fix JSON Schema for string regex with top-level alt. (#9903)

jemc committed 358 days ago

Verified 66c2c930

llama : add tensor name for "result_norm" (#9907)

MollySophia committed 358 days ago

Verified 10433e8b

server : fix the disappearance of the end of the text (#9867)

z80maniac committed 358 days ago

Verified 1f66b699

ggerganov committed 358 days ago

Verified 0e41b300

ggml-alloc : remove buffer_id from leaf_alloc (ggml/987)

danbev committed 358 days ago

Verified cd60b88b

[CANN] Fix cann compilation error (#9891)

leo-pony committed 359 days ago

Verified becfd387

llama : add infill sampler (#9896)

ggerganov committed 359 days ago

Verified 755a9b2b

server : improve infill context reuse (#9894)

ggerganov committed 359 days ago

Verified 223c25a7

sampling : add XTC sampler (#9742)

MaggotHATE committed 359 days ago

Verified fbc98b74

server : update preact (#9895)

ggerganov committed 359 days ago

Verified dcdd5353

readme : update bindings list (#9889)

srgtuszy committed 359 days ago

Verified 4c42f93b

server : handle "logprobs" field with false value (#9871)

VoidIsVoid committed 361 days ago

Verified a89f75e1

Vectorize load instructions in dmmv f16 CUDA kernel (#9816)

agray3 committed 361 days ago

Verified 13dca2a5

server : accept extra_context for the infill endpoint (#9874)

ggerganov committed 361 days ago

Verified d4c19c0f

server : reuse cached context chunks (#9866)

ggerganov committed 361 days ago

Verified c7181bd2

flake.lock: Update (#9870)

ggerganov committed 362 days ago

Verified 92be9f12

server : add option to time limit the generation phase (#9865)

ggerganov committed 362 days ago

Verified edc26566

server : remove self-extend features (#9860)

ggerganov committed 362 days ago

Verified 1bde94dd

server : remove legacy system_prompt feature (#9857)

ggerganov committed 362 days ago

Verified 95c76e8e

llama : improve infill support and special token detection (#9798)

ggerganov committed 363 days ago

Verified 11ac9800

musa : update doc (#9856)

yeahdongcn committed 363 days ago

Verified 943d20b4

ggml : move more prints to the ggml log system (#9839)

slaren committed 363 days ago

Verified 96776405

common : use common_ prefix for common library functions (#9805)

slaren committed 364 days ago

Verified 7eee341b

Older