ggerganov/llama.cpp

Pull Requests Commits
llama : support NUL bytes in tokens
compilade committed 1 year ago
faaac59d
gguf-py : Numpy dequantization for most types (#8939)
compilade committed 1 year ago
Verified 4134999e
flake.lock: Update (#8979)
ggerganov committed 1 year ago
Verified 8cd1bcfd
update guide (#8909)
arthw committed 1 year ago
Verified a21c6fd4
llama : check all graph nodes when searching for result_embd_pooled (#8956)
fairydreaming committed 1 year ago
Verified 33309f66
Optimize Vulkan backend for better CPU performance and less GPU synchronization overhead. (#8943)
mtavenrath committed 1 year ago
Verified 7c5bfd57
metal : fix uninitialized abort_callback (#8968)
slaren committed 1 year ago
Verified 6e02327e
llama : default n_swa for phi-3 (#8931)
ngxson committed 1 year ago
Verified 7eb23840
Add support for encoder-only T5 models (#8900)
fairydreaming committed 1 year ago
Verified 7c3f55c1
gguf-py : fix double call to add_architecture() (#8952)
tarilabs committed 1 year ago
Verified 911b437f
Merge commit from fork
ggerganov committed 1 year ago
Verified b72942fa
llama : add support for lora adapters in T5 model (#8938)
fairydreaming committed 1 year ago
Verified 6afd1a99
make : fix llava obj file race (#8946)
ggerganov committed 1 year ago
Verified 272e3bd9
llama : better replace_all (cont) (#8926)
ggerganov committed 1 year ago
Verified 45a55b91
llava : support MiniCPM-V-2.5 (#7599)
tc-mb committed 1 year ago
Verified 3071c0a5
sync : ggml
ggerganov committed 1 year ago
Verified 4305b57c
whisper : use vulkan as gpu backend when available (whisper/2302)
mstephenson6 committed 1 year ago
Verified 70c0ea35
embedding : add --pooling option to README.md [no ci] (#8934)
danbev committed 1 year ago
Verified 5b2c04f4
llama : fix typo in llama_tensor_get_type comment [no ci] (#8937)
danbev committed 1 year ago
Verified 6f6496bb
server : add one level list nesting for embeddings (#8936)
gelim committed 1 year ago
Verified daef3ab2
llama : reduce useless copies when saving session (#8916)
compilade committed 1 year ago
Verified 345a686d
gguf-py : simplify support for quant types (#8838)
compilade committed 1 year ago
Verified 3a14e003
scripts : sync cann files (#0)
ggerganov committed 1 year ago
Verified afd27f01
scripts : fix sync filenames (#0)
ggerganov committed 1 year ago
Verified 366d486c
sync : ggml
ggerganov committed 1 year ago
Verified e44a561a
ggml : ignore more msvc warnings (ggml/906)
iboB committed 1 year ago
Verified f93d49ab
metal : fix struct name (ggml/912)
ggerganov committed 1 year ago
Verified 5b33ea1e
metal : add abort callback (ggml/905)
conradev committed 1 year ago
Verified 85fca8de
make : clean llamafile objects (#8923)
DrDub committed 1 year ago
Verified ebd541a5
make : use C compiler to build metal embed object (#8899)
slaren committed 1 year ago
Verified 15fa07a5
Terms Of Use Privacy Policy Refund Policy Impressum