Alex Cheema
|
04df31b13b
|
bump up tinygrad to 232edcfd4f8b388807c64fb1817a7668ce27cbad
|
2024-09-25 10:32:49 +01:00 |
|
Alex Cheema
|
93224799e5
|
implement tailscale discovery module
|
2024-09-21 17:55:51 +01:00 |
|
Alex Cheema
|
744b9576f6
|
bump mlx to 0.17.3, bump mlx-lm to 0.18.2
|
2024-09-20 17:56:44 +01:00 |
|
Alex Cheema
|
d917778e2b
|
update mlx to 0.17.1 (not sure where 0.17.0 went on PyPi disappeared)g
|
2024-08-25 22:20:27 +01:00 |
|
Alex Cheema
|
09a8468395
|
upgrade mlx to 0.17.0
|
2024-08-23 22:13:24 +01:00 |
|
Alex Cheema
|
2e27076665
|
simplify formatting with yapf
|
2024-08-22 14:04:28 +01:00 |
|
Alex Cheema
|
cea9b48d24
|
update mlx-lm to 0.17.0, use lru caches for kv_cache with RotatingKVCache to optimise memory fixes #158
|
2024-08-20 17:33:36 +01:00 |
|
Alex Cheema
|
92dbb3204d
|
update mlx to 0.16.3
|
2024-08-15 22:23:18 +01:00 |
|
Alex Cheema
|
440fd35ea7
|
upgrade aiohttp
|
2024-08-10 14:32:51 +01:00 |
|
Alex Cheema
|
71591d2ebc
|
display all interfaces web chat and chatgpt api are available on fixes #134
|
2024-08-09 13:32:55 +01:00 |
|
Alex Cheema
|
35b7042e70
|
upgrade mlx to 0.16.1
|
2024-08-07 14:19:59 +01:00 |
|
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
|
2024-08-05 23:03:17 +01:00 |
|
Alex Cheema
|
d6a7e46324
|
async model downloading with download progress. fixes #102. related: #16 #104
|
2024-07-31 22:47:03 +01:00 |
|
Alex Cheema
|
78db451d7e
|
add pillow to main dependencies
|
2024-07-30 14:27:45 +01:00 |
|
Alex Cheema
|
824f05263f
|
Merge branch 'main' into HEAD
|
2024-07-30 14:18:10 +01:00 |
|
Alex Cheema
|
142682645f
|
bump up tinygrad version
|
2024-07-30 10:02:20 +01:00 |
|
Varshith
|
acc94b50c7
|
chatgpt api integration
|
2024-07-28 16:12:21 +05:30 |
|
Alex Cheema
|
b44b917151
|
add pillow as testing dependency
|
2024-07-27 20:23:03 -07:00 |
|
Alex Cheema
|
ce761038ac
|
formatting / linting
|
2024-07-27 17:01:37 -07:00 |
|
Alex Cheema
|
bbfd5adc20
|
add support for llama3.1 (8b, 70b, 405b). bump mlx up to 0.16.0 and mlx-lm up to 0.16.1. fixes #66
|
2024-07-23 14:41:34 -07:00 |
|
Alex Cheema
|
4e46232364
|
add simple prometheus metrics collection, with a prometheus / grafana instance for live dashboard. related: #22
|
2024-07-22 02:38:37 -07:00 |
|
Alex Cheema
|
4b592f9d45
|
exo topology visualisation that shows the topology of the network, device capabilities and the currently active node using opaque statuses. fixes #36. ready for #33
|
2024-07-18 22:50:18 -07:00 |
|
Alex Cheema
|
46d618abed
|
tiny fixes
|
2024-07-18 04:33:44 -07:00 |
|
Alex Cheema
|
071b1caa0b
|
drop exo to 0.0.1 (still experimental)
|
2024-07-18 01:17:19 -07:00 |
|
Alex Cheema
|
8762effaf4
|
chatgpt api repsonse streaming solves #20
|
2024-07-17 19:00:18 -07:00 |
|
Alex Cheema
|
998d484384
|
match psutil platform detection might catch some edge cases
|
2024-07-17 11:48:52 -07:00 |
|
Alex
|
bfaeccc7d5
|
added setup py
|
2024-07-17 13:21:46 +02:00 |
|