ochafik/llama.cpp

Pull Requests Commits
ggml : restore ggml_rope_xpos_inplace (ggml/0)
ggerganov committed 2 years ago
Verified 72de268b
Add Arc A750 and Arch linux to readme-sycl.md as verified GPU model and Linux distro (#7605)
Akarshan Biswas committed 2 years ago
Verified 0e8d8bfd
ggml : fix typo in ggml.c (#7603)
zhouwg committed 2 years ago
Verified 504f0c34
[SYCL] Align GEMM dispatch (#7566)
airMeng committed 2 years ago
Verified b864b50c
Tokenizer WPM fixes (#7500)
jaime-m-p committed 2 years ago
Verified 02c1ecad
sycl : fix assert (#7563)
ggerganov committed 2 years ago
Verified 6bd12ce4
llama : support small Granite models (#7481)
giuseppe committed 2 years ago
Verified 5442939f
vulkan: properly initialize vulkan devices for LLAMA_SPLIT_MODE_NONE (#7552)
Adriankhl committed 2 years ago
Verified 56411a95
rpc : resource management rework (#7562)
rgerganov committed 2 years ago
Verified 2b737caa
Add support for DeepseekV2ForCausalLM (#7519)
fairydreaming committed 2 years ago
Verified ee3dff6b
tests : fix test-tokenizer-0.sh
ggerganov committed 2 years ago
Verified edc29433
llama : handle unknown utf8 bytes (#7588)
ggerganov committed 2 years ago
Verified 8b99e2aa
github: add refactor to issue template (#7561)
mofosyne committed 2 years ago
Verified 271ff3fc
[SYCL]fix ggml_sycl_mul_mat_id() to match the change of api (#7436)
arthw committed 2 years ago
Verified e2b06507
ggml : generalize GGML_OP_CONCAT (#7563)
ggerganov committed 2 years ago
Verified 0548a418
server: do not remove whitespace at the start of a completion chunk (#7524)
mgroeber9110 committed 2 years ago
Verified 9335b969
Markdownish code block fix (#7571)
nathan-sixnines committed 2 years ago
Verified c4176715
llava : update clip.h (#7580)
eltociear committed 2 years ago
Verified 74b239b3
update HIP_UMA #7399 (#7414)
Djip007 committed 2 years ago
Verified 852aafb1
adding in x64 targets to cmake presets (#7574)
kunnis committed 2 years ago
Verified 0136966d
make: add --device-debug to NVCC debug flags (#7542)
JohannesGaessler committed 2 years ago
Verified 10b1e458
Allow multiple copy function pointers for CUDA graph kernel param updates (#7565)
agray3 committed 2 years ago
Verified 197c0068
Fix q_xxs using mul_mat_q (#7459)
AidanBeltonS committed 2 years ago
Verified 95f84d5c
Add freq factors (#7495)
AidanBeltonS committed 2 years ago
Verified 5487593b
metal : add GGML_OP_REPEAT kernels (#7557)
ggerganov committed 2 years ago
Verified 1d8fca72
metal : disable FA kernel for HS=256 (#7556)
ggerganov committed 2 years ago
Verified 62bfef51
llama : add comments about experimental flags (#7544)
ggerganov committed 2 years ago
Verified eaf6e031
github: add self sorted issue ticket forms (#7543)
mofosyne committed 2 years ago
Verified d6ef0e77
flake.lock: Update (#7540)
ggerganov committed 2 years ago
Verified dff451cf
main: replace --no-special with --special (#7534)
mofosyne committed 2 years ago
Verified d298382a
Terms Of Use Privacy Policy Refund Policy Impressum