mlhq-kg-mac

Project for builiing Agent Flow with Knowledge Graphs on MAC M4 Pro (and some API stuff)

NOTE Most of the inspiration for this project came from here (https://github.com/PeterGriffinJin/Graph-CoT/tree/main) and some of the code related to GraphAgent were taken from there - but updated for my needs.

Follow-ups

QUES: How to route my HuggingFace InferenceClient inference to the MLHQ org acount rather than my own?

https://www.reddit.com/r/huggingface/comments/1idkjx5/hf_new_inference_providers_pricing_confusion/

https://huggingface.co/docs/api-inference/en/pricing

Current Issues:

1. Huggingface InferenceClient does not support local execution

• Therefore I could either create an account and use their infra or
• I could build my own wrapper for the support.

^^^ For now, I am going with the first option

Note

For this specific project, I think I can forgo the requirement of OpenAI-API chat completion capatibility. As long as I know the prompt, I can simply push that into the locally hosted LLM via HugginFace Pipelines.

Cost:

So far my general costs are:

• M4 Max 128G → 5k
• HF Entrprise → 216 (billed yearly at 18 a month )
• Claude Pro → 20 bucks a month
• OpenAI Businsess → 600 for a year.