ngxson/llama.cpp

Pull Requests Commits
common : fix duplicated file name with hf_repo and hf_file
ngxson committed 1 year ago
91fd3226
Add some minimal optimizations for CDNA (#10498)
IMbackK committed 1 year ago
Verified 3ad5451f
ci : faster CUDA toolkit installation method and use ccache (#10537)
slaren committed 1 year ago
Verified 46c69e0e
metal : fix group_norm support condition (#0)
ggerganov committed 1 year ago
Verified 9e2301f4
sync : ggml
ggerganov committed 1 year ago
Verified fee824a1
Do not include arm_neon.h when compiling CUDA code (ggml/1028)
frankier committed 1 year ago
Verified 9150f8fe
vulkan: define all quant data structures in types.comp (#10440)
jeffbolznv committed 1 year ago
Verified c31ed2ab
vulkan: Handle GPUs with less shared memory (#10468)
jeffbolznv committed 1 year ago
Verified 5b3466be
vulkan: further optimize q5_k mul_mat_vec (#10479)
jeffbolznv committed 1 year ago
Verified 249a7902
vulkan: skip integer div/mod in get_offsets for batch_idx==0 (#10506)
jeffbolznv committed 1 year ago
Verified 71a64989
vulkan: optimize Q2_K and Q3_K mul_mat_vec (#10459)
jeffbolznv committed 1 year ago
Verified 4a57d362
ci : fix cuda releases (#10532)
slaren committed 1 year ago
Verified c9b00a70
Add OLMo 2 model in docs (#10530)
2015aroras committed 1 year ago
Verified de509735
ci : remove nix workflows (#10526)
slaren committed 1 year ago
Verified 5a349f28
llama : disable warnings for 3rd party sha1 dependency (#10527)
slaren committed 1 year ago
Verified 30ec3983
Fix HIP flag inconsistency & build docs (#10524)
tristandruyen committed 1 year ago
Verified be0e350c
mtgpu: Add MUSA_DOCKER_ARCH in Dockerfiles && update cmake and make (#10516)
yeahdongcn committed 1 year ago
Verified 249cd93d
vulkan: fix group_norm (#10496)
jeffbolznv committed 1 year ago
Verified 904109ed
server : replace behave with pytest (#10416)
ngxson committed 1 year ago
Verified 45abe0f7
restore the condistion to build & update pacakge when merge (#10507)
NeoZhangJianyu committed 1 year ago
Verified 0bbd2262
cmake : enable warnings in llama (#10474)
ggerganov committed 1 year ago
Verified ab96610b
ci : publish the docker images created during scheduled runs (#10515)
slaren committed 1 year ago
Verified 7db3846a
ci : add ubuntu cuda build, build with one arch on windows (#10456)
slaren committed 1 year ago
Verified c6807b3f
ggml-cpu: cmake add arm64 cpu feature check for macos (#10487)
chaxu01 committed 1 year ago
Verified 25669aa9
server : fix parallel speculative decoding (#10513)
ggerganov committed 1 year ago
Verified 84e1c33c
speculative : simplify the implementation (#10504)
ggerganov committed 1 year ago
Verified 811872a5
CANN: Improve the Inferencing Performance for Ascend NPU Device (#10454)
shen-shanshan committed 1 year ago
Verified 9a4b79bc
CANN: RoPE and CANCAT operator optimization (#10488)
noemotiovon committed 1 year ago
Verified 7066b4cc
vulkan: Fix a vulkan-shaders-gen arugment parsing error (#10484)
sparkleholic committed 1 year ago
Verified 0eb4e12b
Introduce llama-run (#10291)
ericcurtin committed 1 year ago
Verified 0cc63754
Terms Of Use Privacy Policy Refund Policy Impressum