The one about our vision and next step

[Aisuko]

We want to work on the AI technique and try to make the LLVMs more specific for our vision's task. So, I believe we can have a simple roadmap here.

• The LoRA technique is a good starting point.
• Maybe we need to figure out which open-source model is more effective for us.
• Re-training it by using promising techniques but are not limited to LoRA.
  • And I want to do the re-training model task on the Steam Deck. We are all expected by using fewer resources to achieve more tasks.
• Writing a specific version of the copilot for myself or our team and it will be released as VSCode's plugin. It should be easy for me and others to access.
• After finishing above, it is time to make some effort on the GNU Hurd.

[Aisuko]

@SkywardAI/contributors Any thoughts here?

[Aisuko]

out of date