roadmap: Built-in Model Library

[dan-menlo]

Model Library

• Work with eng team to update URLs
• Update READMEs
• Provide more quantizations (e.g. full llama.cpp quantizations)
• Document how to create/add a Model Library
• docs: Show contributors how to add model compilations to HF Cortex curated models cortex.so#78
• [TBD] Shift Built-in model repos back to Jan's huggingface (better brand recognition, we are also not pushing Cortex as a brand for now)

Models

• model: Ichigo-llama3.1 #17
• model: llama3.1 #16
• model: Mistral Nemo #19
• model: llama3.2 #26

[dan-menlo]

We will likely need to go over all models, and create a main branch (ie default download)

janhq/cortex.cpp#1076 (comment)