 Daniel Hiltgen
|
8de8729e35
Remove llama.cpp submodule and shift new build to top
|
6 meses atrás |
 Patrick Devine
|
c7cb0f0602
image processing for llama3.2 (#6963)
|
6 meses atrás |
 Daniel Hiltgen
|
f86d00cd95
llama: add compiler tags for cpu features (#7137)
|
6 meses atrás |
 Gabe Goodhart
|
f2890a4494
IBM granite/granitemoe architecture support (#6760)
|
6 meses atrás |
|
Daniel Hiltgen
|
7d6eb0d4c3
Move macos v11 support flags to build script (#7203)
|
6 meses atrás |
|
Daniel Hiltgen
|
5dd0477fd4
Fix regression on older macos versions (#7192)
|
6 meses atrás |
 Jeffrey Morgan
|
96efd9052f
Re-introduce the `llama` package (#5034)
|
6 meses atrás |
 Michael Yang
|
fccf8d179f
partial decode ggml bin for more info
|
1 ano atrás |
 Bruce MacDonald
|
984c9c628c
fix embeddings invalid values
|
1 ano atrás |
|
Bruce MacDonald
|
09d8bf6730
fix build errors
|
1 ano atrás |
|
Bruce MacDonald
|
7a5f3616fd
embed text document in modelfile
|
1 ano atrás |
|
Michael Yang
|
f2074ed4c0
Merge pull request #306 from jmorganca/default-keep-system
|
1 ano atrás |
|
Bruce MacDonald
|
a6f6d18f83
embed text document in modelfile
|
1 ano atrás |
|
Jeffrey Morgan
|
5eb712f962
trim whitespace before checking stop conditions
|
1 ano atrás |
|
Michael Yang
|
4dc5b117dd
automatically set num_keep if num_keep < 0
|
1 ano atrás |
|
Michael Yang
|
b9f4d67554
configurable rope frequency parameters
|
1 ano atrás |
|
Michael Yang
|
c5bcf32823
update llama.cpp
|
1 ano atrás |
|
Michael Yang
|
74a5f7e698
no gpu for 70B model
|
1 ano atrás |
|
Michael Yang
|
319f078dd9
remove -Werror
|
1 ano atrás |
|
Jeffrey Morgan
|
7da249fcc1
only build metal for `darwin,arm` target
|
1 ano atrás |
|
Bruce MacDonald
|
184ad8f057
allow specifying stop conditions in modelfile
|
1 ano atrás |
|
Michael Yang
|
3549676678
embed ggml-metal.metal
|
1 ano atrás |
|
Michael Yang
|
fadf75f99d
add stop conditions
|
1 ano atrás |
|
Michael Yang
|
ad3a7d0e2c
add NumGQA
|
1 ano atrás |
|
Michael Yang
|
cca61181cb
sample metrics
|
1 ano atrás |
|
Michael Yang
|
c490416189
lock on llm.lock(); decrease batch size
|
1 ano atrás |
|
Michael Yang
|
f62a882760
add session expiration
|
1 ano atrás |
|
Michael Yang
|
3003fc03fc
update predict code
|
1 ano atrás |
|
Michael Yang
|
35af37a2cb
session id
|
1 ano atrás |
|
Michael Yang
|
726bc647b2
enable k quants
|
1 ano atrás |
