ggml-org/ggml

Pull Requests Commits
ci : test push into a different branch
ggerganov committed 2 years ago
Verified c5d62d8c
ci : fix
ggerganov committed 2 years ago
Verified a1caff0f
ci : move out the README.md header generation
ggerganov committed 2 years ago
Verified 335f98d0
ci : utilize gg_printf
ggerganov committed 2 years ago
Verified 185ce86a
ci : fix header
ggerganov committed 2 years ago
Verified 9fa43699
ci : update README.md
ggerganov committed 2 years ago
Verified 7742df9d
ci : generate README.md
ggerganov committed 2 years ago
Verified 97c85ce3
ci : disable test-opt for release as well
ggerganov committed 2 years ago
Verified d77c4468
ci : output time of commands
ggerganov committed 2 years ago
Verified e7de20a1
ci : try fix pipefail
ggerganov committed 2 years ago
Verified a5247db3
ci : try to capture exit codes
ggerganov committed 2 years ago
Verified 09e1f917
ci : induce error in ci-1
ggerganov committed 2 years ago
Verified a588a8a5
ci : fix ctest
ggerganov committed 2 years ago
Verified aa4f9c6e
ci : ignore test-opt
ggerganov committed 2 years ago
Verified 7f4fae64
ci : remove deps
ggerganov committed 2 years ago
Verified b54f8dca
run : fix
ggerganov committed 2 years ago
Verified 87be9305
run : add deps
ggerganov committed 2 years ago
Verified f95d3452
ci : remove setup.sh
ggerganov committed 2 years ago
Verified 7851095d
ci : add initial scripts
ggerganov committed 2 years ago
Verified 7fe24f07
ggml : sync llama.cpp (CUDA add/mul bcast + Metal fix + custom RoPE)
ggerganov committed 2 years ago
Verified 0c169489
starcoder : mmap (and gpu) example (#338)
johnson442 committed 2 years ago
Verified d2b178ee
cuda : fix CUDA arch: OFF -> 52;61 (#389)
JohannesGaessler committed 2 years ago
Verified 90cb1ef6
ggml : sync llama.cpp (fix CUDA build, faster Metal)
ggerganov committed 2 years ago
Verified 5bef75b1
cuda : support GLM-style RoPE (#383)
li-plus committed 2 years ago
Verified da3015e5
ggml : retire custom ggml_2d impl in favor of general (#352)
ggerganov committed 2 years ago
Verified d7179d77
ggml : general conv_2d CPU implementation (#352)
dmille committed 2 years ago
Verified a1bd9826
ggml : fix mul_mat src1 indexing when not contiguous and non-F32 (#386)
ggerganov committed 2 years ago
Verified 64d645ab
ggml : fix mul_mat src1 indexing when src1 is not contiguous (#386)
ggerganov committed 2 years ago
Verified e55e3473
ggml : apply mul_mat broadcast fix (sync llama.cpp)
ggerganov committed 2 years ago
Verified f6365c06
cuda : sync llama.cpp
ggerganov committed 2 years ago
Verified 56216523
Terms Of Use Privacy Policy Refund Policy Impressum