 Alex Cheema
|
83d4d2b3d8
update mlx to 0.20.0, mlx-lm to 0.19.3
|
5 mēneši atpakaļ |
|
Alex Cheema
|
bc7acfd37b
update mlx to 0.19.3, mlx-lm to 0.19.2
|
6 mēneši atpakaļ |
|
Alex Cheema
|
f82410f808
clear instructions on formatting with yapf, remove linting
|
6 mēneši atpakaļ |
 Alex Cheema
|
03621b9962
Merge pull request #368 from ianpaul10/feat/manual-disc-0
|
6 mēneši atpakaļ |
 willschneider15
|
ae47fe11e6
Moving conditonal apple silicon logic to setup.py
|
6 mēneši atpakaļ |
|
willschneider15
|
e1ae6d5a4f
Getting the setup script to work on Intel based Mac machines
|
6 mēneši atpakaļ |
 Ian Paul
|
6b48a936b9
add pydantic dependency
|
6 mēneši atpakaļ |
|
Alex Cheema
|
7b2a523fd1
Merge pull request #289 from GaetanLepage/pynvml
|
7 mēneši atpakaļ |
|
Alex Cheema
|
e8a8702377
replace tailscale.devices with good old http, removing the need for tailscale dependency
|
7 mēneši atpakaļ |
 Gaetan Lepage
|
9ffd8116f0
Use official nvidia-ml-py instead of pynvml
|
7 mēneši atpakaļ |
|
Alex Cheema
|
552c04fb49
add back Jinja2
|
7 mēneši atpakaļ |
|
Alex Cheema
|
b331592647
remove blobfile, tiktoken, tokenizers
|
7 mēneši atpakaļ |
|
Alex Cheema
|
413ecb1b12
remove hf-transfer, huggingface-hub, Jinja2 unused dependencies
|
7 mēneši atpakaļ |
|
Gaetan Lepage
|
c5b38f450f
move tinychat inside exo package
|
7 mēneši atpakaļ |
|
Gaetan Lepage
|
fa67ee9bba
add entry_point exo to run main.py
|
7 mēneši atpakaļ |
|
Alex Cheema
|
3e13e5ed6a
upgrade mlx to 0.18.0
|
7 mēneši atpakaļ |
|
Alex Cheema
|
04df31b13b
bump up tinygrad to 232edcfd4f8b388807c64fb1817a7668ce27cbad
|
7 mēneši atpakaļ |
|
Alex Cheema
|
93224799e5
implement tailscale discovery module
|
7 mēneši atpakaļ |
|
Alex Cheema
|
744b9576f6
bump mlx to 0.17.3, bump mlx-lm to 0.18.2
|
7 mēneši atpakaļ |
|
Alex Cheema
|
d917778e2b
update mlx to 0.17.1 (not sure where 0.17.0 went on PyPi disappeared)g
|
8 mēneši atpakaļ |
|
Alex Cheema
|
09a8468395
upgrade mlx to 0.17.0
|
8 mēneši atpakaļ |
|
Alex Cheema
|
2e27076665
simplify formatting with yapf
|
8 mēneši atpakaļ |
|
Alex Cheema
|
cea9b48d24
update mlx-lm to 0.17.0, use lru caches for kv_cache with RotatingKVCache to optimise memory fixes #158
|
8 mēneši atpakaļ |
|
Alex Cheema
|
92dbb3204d
update mlx to 0.16.3
|
8 mēneši atpakaļ |
|
Alex Cheema
|
440fd35ea7
upgrade aiohttp
|
8 mēneši atpakaļ |
|
Alex Cheema
|
71591d2ebc
display all interfaces web chat and chatgpt api are available on fixes #134
|
8 mēneši atpakaļ |
|
Alex Cheema
|
35b7042e70
upgrade mlx to 0.16.1
|
9 mēneši atpakaļ |
|
Alex Cheema
|
545a486ed3
separate hf_helpers, make extra dir with download_hf script, unify downloading so tinygrad uses the same method as mlx and interoperable model formats
|
9 mēneši atpakaļ |
|
Alex Cheema
|
d6a7e46324
async model downloading with download progress. fixes #102. related: #16 #104
|
9 mēneši atpakaļ |
|
Alex Cheema
|
78db451d7e
add pillow to main dependencies
|
9 mēneši atpakaļ |
