Evan Quiney

Support all MLX backends

2

MLX supports many computational backends - cpu, cuda, metal and gpu. It should be easy to specify this information as a parallel part of the InstanceMeta, and pass it into...

This is an extension of #964 with some cleanup.

## Describe the bug In the dashboard, the minimum model size is 4GB. This makes it impossible to attempt to load smaller models on devices 4GB of memory, and often...

Separate out the Runner's behaviour into a "connect" phase and a "load" phase

1

## Motivation We should ensure all runners are connected before loading the model - this gives us finer grained control in the future for the workers planning mechanism over the...