So JiRack LLM delivered in Onnx format by AI community recommendations and modern LLM standards .
it uses huggingface tokenizer fr the onnx model - tokenizer.json
Life demo the model on AWS http://www.cmsmanhattan.com/Productlist.jsp?catalog_id=-2#
Youtube Demo: https://www.youtube.com/watch?v=vHClQu76kMc
It is working on ONNX Runtime
Please use to check JiRack LLM architure to work So foolow the link and open JiRack LLM https://netron.app/
Video how deploy JiRack in RAG System
RAG repo with deployment scripts . So use Git and Docker to install JiRack and CMS Manhattan RAG System git clone https://grabko1@bitbucket.org/cmsmanhattan/rag.git
Watch RAG System deployment on any docker cloud https://www.youtube.com/watch?v=M4Q8_Dr35Cc
Cms Manhattan research foe new GPT model . New GPT mode interview with DJL on java https://youtu.be/sFXTL0g875s
Follow to your future