74eb4678ad
fix error handling in LlamaEvent
...
also fixes #4
2023-05-03 09:02:54 +01:00
616a2b9dc9
add note about CameLLM to README
2023-04-26 13:43:13 +01:00
ed25d79c76
fix build.yml
2023-04-14 12:03:44 +01:00
604a1a9ec4
add notes about v2 to the README
2023-04-14 12:01:08 +01:00
140fc89987
Create CODE_OF_CONDUCT.md
2023-04-11 13:56:19 +01:00
efa3088bc6
update README w/ API changes and additional detail
2023-03-18 00:02:16 +01:00
726f2b2ca1
set deployment targets in Package.swift
2023-03-17 20:49:40 +01:00
728b08163c
use async API in llamaTest
2023-03-17 20:45:48 +01:00
8b04fd8250
add async version of run()
2023-03-17 20:40:07 +01:00
4e1b8519a1
set deployment targets to macOS 10.15 and iOS 13.0
2023-03-17 20:32:59 +01:00
1e3ce76948
add more details to README and tidy up
2023-03-16 03:06:57 +01:00
28e35a14d3
move convert-pth-to-ggml to tools/
2023-03-16 02:43:22 +01:00
a58bc91240
remove quantize binary from repo and ensure it is ignored
2023-03-16 02:38:56 +01:00
eaef7a35d6
fix quantize models path
2023-03-16 02:38:19 +01:00
443fe5d338
exclude quantize.cpp from llamaObjCxx
2023-03-16 02:31:41 +01:00
713b65df3e
add back Makefile and quantize.cpp in correct locations
2023-03-16 02:30:13 +01:00
e0d648c817
remove quantize.cpp
2023-03-16 01:30:35 +01:00
058377cb3e
add SPM support
2023-03-16 01:17:44 +01:00
5f4905a256
separate Swift and C++/Objective-C++ code into separate libraries
2023-03-16 00:24:12 +01:00
b44026b8db
remove Makefile and CMakeLists.txt
...
will pull in any changes from llama.cpp and adjust the Xcode configuration
to match this.
2023-03-15 10:49:00 +01:00
56c19351db
Merge remote-tracking branch 'llama.cpp/master'
2023-03-15 10:36:15 +01:00
47857e564c
Don't use vdotq_s32 if it's not available ( #139 )
...
* Don't use vdotq_s32 if it's not available
`dotprod` extensions aren't available on some ARM CPUs (e.g. Raspberry Pi 4), so check for them and only use them if they're available.
Reintroduces the code removed in 84d9015
2023-03-14 21:34:37 +02:00
60f819a2b1
Add section to README on how to run the project on Android ( #130 )
2023-03-14 15:30:08 +02:00
8ce62d4135
improve README
2023-03-14 12:03:12 +01:00
27afd915b8
update license file
2023-03-14 11:53:49 +01:00
f5779567ed
update readme
2023-03-14 11:52:34 +01:00
e69e116b82
improve output
2023-03-14 11:45:59 +01:00
e99e6f1fbf
make sample app interactive
2023-03-14 11:45:59 +01:00
81e97c8585
remove prints to stderr and return NSErrors
2023-03-14 11:45:59 +01:00
2b27f14035
remove debug printf() statements
2023-03-14 11:45:59 +01:00
60458cc580
improve swift API and test app
2023-03-14 11:45:59 +01:00
e3df7bb91f
add basic error handling
2023-03-14 11:45:59 +01:00
e04b99370c
pass events from C++ through to Swift
2023-03-14 11:45:59 +01:00
b3404f6d27
execute prediction in NSOperation
2023-03-14 11:45:59 +01:00
95722fb0e9
support reverse prompt in config
2023-03-14 11:45:59 +01:00
9e54dc6d56
wire up LlamaRunner to cpp code
2023-03-14 11:45:59 +01:00
406470dd3b
move cpp files to separate directory
2023-03-14 11:45:59 +01:00
f33e7dbed3
fix file check in llamaTest
2023-03-14 11:45:59 +01:00
2717aee3c1
add basic Objective-C runner and pass model path
2023-03-14 11:45:59 +01:00
7e27bbc02e
add basic Xcode project and include cpp files
2023-03-14 11:45:59 +01:00
97ab2b2578
Add Misc section + update hot topics + minor fixes
2023-03-14 09:43:52 +02:00
2f700a2738
Add windows to the CI ( #98 )
2023-03-13 22:29:10 +02:00
c09a9cfb06
CMake build in Release by default ( #75 )
2023-03-13 21:22:15 +02:00
7ec903d3c1
Update contribution section, hot topics, limitations, etc.
2023-03-13 19:21:51 +02:00
4497ad819c
Print system information
2023-03-13 19:15:08 +02:00
ed6849cc07
Initial support for CMake ( #75 )
2023-03-13 19:12:33 +02:00
41be0a3b3d
Add NetBSD support. ( #90 )
2023-03-13 18:40:54 +02:00
671d5cac15
Use fprintf for diagnostic output ( #48 )
...
keep printf only for printing model output
one can now use ./main ... 2>dev/null to suppress any diagnostic output
2023-03-13 18:39:56 +02:00
84d9015c4a
Use vdotq_s32 to improve performance ( #67 )
...
* 10% performance boost on ARM
* Back to original change
2023-03-13 18:36:44 +02:00
63fd76fbb0
Reduce model loading time ( #43 )
...
* Use buffering
* Use vector
* Minor
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2023-03-13 18:33:43 +02:00
Alex Rozanski