Commits

History for

Commits on Jun 26, 2023

k-quants : support for super-block size of 64 (ggerganov#2001 )

ikawrakow
and
Kawrakow
committedJun 26, 2023

Commits on Jun 18, 2023

metal : handle buffers larger than device's maxBufferLength (ggerganov#1826 )
ggerganov
committedJun 18, 2023

Commits on Jun 17, 2023

minor : warning fixes
ggerganov
committedJun 17, 2023
metal : add norm, cpy f16->f16, alibi kernels (ggerganov#1823 )
apage43
committedJun 17, 2023

Commits on Jun 15, 2023

metal : parallel command buffer encoding (ggerganov#1860 )
ggerganov
committedJun 15, 2023

Commits on Jun 12, 2023

Metal implementation for all k_quants (ggerganov#1807 )

ikawrakow
and
Kawrakow
committedJun 12, 2023
metal : fix failure to load model (ggerganov#1817 )

ikawrakow
and
Kawrakow
committedJun 12, 2023

Commits on Jun 10, 2023

metal : fix issue with ggml-metal.metal path. Closes ggerganov#1769 (ggerganov#1782 )
abetlen
committedJun 10, 2023
metal : add Q4_1 implementation (ggerganov#1785 )

ikawrakow
and
Kawrakow
committedJun 10, 2023

Commits on Jun 9, 2023

metal : add GELU implementation (ggerganov#1770 )

manyoso
and
Adam Treat
committedJun 9, 2023
metal : faster q4_0 (ggerganov#1775 )

ikawrakow
and
Kawrakow
committedJun 9, 2023

Commits on Jun 8, 2023

metal : add Q2_K implementation (ggerganov#1762 )

ikawrakow
and
Kawrakow
committedJun 8, 2023
metal : Q6_K implementation (ggerganov#1752 )

ikawrakow
and
Kawrakow
committedJun 8, 2023
metal : add Q4_K implementation (ggerganov#1733 )

ikawrakow
and
Kawrakow
committedJun 8, 2023

Commits on Jun 6, 2023

metal : add f16 support
ggerganov
committedJun 6, 2023
metal : add checks for buffer size (ggerganov#1706 )

spencersutton
and
Spencer Sutton
committedJun 6, 2023

Commits on Jun 5, 2023

metal : use shared buffers between CPU and GPU (ggerganov#1696 )

kiltyj
and
ggerganov
committedJun 5, 2023

Commits on Jun 4, 2023

llama : Metal inference (ggerganov#1642 )
ggerganov
committedJun 4, 2023

Loading