Mayank Chhabra
|
9d086c2e28
|
Lock 13B and 70B models in memory for faster interference
|
2023-08-18 03:11:42 +07:00 |
Mayank Chhabra
|
2a24cb9e60
|
Set 70B's GQA to 8
|
2023-08-18 00:18:19 +07:00 |
Mayank Chhabra
|
bc1300fba5
|
Update docker service names for different model APIs
|
2023-08-17 23:28:55 +07:00 |
Mayank Chhabra
|
bb8e5fccaa
|
Add restart on-failure policy
|
2023-08-17 04:17:47 +07:00 |
Mayank Chhabra
|
680ff5144f
|
Increase api ping timeout to 10 min
|
2023-08-16 01:27:51 +07:00 |
Mayank Chhabra
|
8907e13a3e
|
Comment out 70B image builds
|
2023-08-16 00:32:08 +07:00 |
Mayank Chhabra
|
b0b059a05a
|
Wait for api to be available before starting UI
|
2023-08-15 23:42:08 +07:00 |
Mayank Chhabra
|
ee97955bb7
|
Add support for 13B and 70B models, workflow, readme
|
2023-08-15 23:11:39 +07:00 |