Welcome to the llmware HuggingFace page. We believe that the ascendence of LLMs creates a major new application pattern and data pipelines that will be transformative in the enterprise, especially in knowledge-intensive industries. Our open source research efforts are focused both on the new "ware" ("middleware" and "software" that will wrap and integrate LLMs), as well as building high-quality automation-focused enterprise Agent, RAG and embedding small specialized language models.
Our model training initiatives fall into four major categories:
SLIMs - small, specialized function calling models for stacking in multi-model, Agent-based workflows -- SLIMs
BLING/DRAGON - highly-accurate fact-based question-answering models
-- small model accuracy benchmark |
our journey building small accurate language models
Industry-BERT - industry fine-tuned embedding models
Private Inference - Self-Hosting, Packaging and Quantization - GGUF, ONNX, OpenVino
Please check out a few of our recent blog postings related to these initiatives:
thinking does not happen one token at a time |
rag instruct test dataset |
llmware emerging stack |
becoming a master finetuning chef
Interested? Join us on Discord