ggml-org/llama.cpp

Pull Requests Commits
readme : add tenere in the ui tools list (#5284)
pythops committed 2 years ago
Verified 6a66c507
Fix im2col with 32fp (#5286)
AidanBeltonS committed 2 years ago
Verified a305dba8
perplexity : fix KL divergence calculations on Windows (#5273)
kalomaze committed 2 years ago
Verified 19122117
scripts : parse wtype in server-llm.sh (#5167)
ggerganov committed 2 years ago
Verified e437b37f
py : add check for '.attn.masked_bias' layers to GPT2model (#5281)
MirrorAzure committed 2 years ago
Verified 2d40085c
Tidy ggml-sycl (#5261)
AidanBeltonS committed 2 years ago
Verified b05102fe
docker : add build for SYCL, Vulkan + update readme (#5228)
ngxson committed 2 years ago
Verified 6b91b1e0
[SYCL] get MAX_MEM_ALLOC from device property (#5270)
airMeng committed 2 years ago
Verified e805f0fa
[SYCL] update guide of SYCL backend (#5254)
NeoZhangJianyu committed 2 years ago
Verified af3ba5d9
llama : fix memory leak in llama_batch_free (#5252)
irbull committed 2 years ago
Verified e1e72109
add --no-mmap in llama-bench (#5257)
NeoZhangJianyu committed 2 years ago
Verified 128dcbd3
Vulkan Phi Fix for AMD Proprietary Drivers (#5260)
0cc4m committed 2 years ago
Verified 4d0924a8
cuda : fix LLAMA_CUDA_F16 (#5262)
slaren committed 2 years ago
Verified 8ca511ca
make : generate .a library for static linking (#5205)
nehzata committed 2 years ago
Verified d71ac909
llama : support InternLM2 (#5184)
SolenoidWGT committed 2 years ago
Verified ce320601
Fix broken Vulkan Cmake (properly) (#5230)
netrunnereve committed 2 years ago
Verified 1cfb5372
llama : reorder build_orion() at correct place (#5118)
ggerganov committed 2 years ago
Verified d3bac7d5
llama : remove LLAMA_MAX_DEVICES and LLAMA_SUPPORTS_GPU_OFFLOAD (#5240)
ggerganov committed 2 years ago
Verified 5cb04dbc
metal : add im2col F32 dst support (#5132)
ggerganov committed 2 years ago
Verified efb7bdbb
llava : add MobileVLM support (#5132)
JidongZhang-THU committed 2 years ago
Verified 15606309
format license text, restore apache license by legal suggestion (#5233)
NeoZhangJianyu committed 2 years ago
Verified b2b9f025
ggml : limit n_threads to the max n_tasks (#5238)
slaren committed 2 years ago
Verified dabcc5b4
Vulkan Fixes (#5223)
0cc4m committed 2 years ago
Verified f8e9140c
Fix typos of IQ2_XXS and IQ3_XXS in llama.cpp (#5231)
ymcui committed 2 years ago
Verified d62520eb
support SYCL backend windows build (#5208)
NeoZhangJianyu committed 2 years ago
Verified 01684139
kompute : llama-bench support and ggml_cpu_has_kompute() (#5226)
cebtenzzre committed 2 years ago
Verified e8dc55d0
Revert "server : change deps.sh xxd files to string literals (#5221)"
ggerganov committed 2 years ago
Verified e0085fdf
server : fix context shift (#5195)
ggerganov committed 2 years ago
Verified e6f291d1
server : change deps.sh xxd files to string literals (#5221)
jboero committed 2 years ago
Verified 4003be0e
ggml : fix IQ3_XXS on Metal (#5219)
ikawrakow committed 2 years ago
Verified fea4fd4b
Terms Of Use Privacy Policy Refund Policy Impressum