slim-boolean-phi-3-onnx

slim-boolean-phi-3-onnx is a specialized function calling model optimized for boolean (yes/no) questions. The model expects as input a text passage context, and a boolean question, and generates a python dictionary consisting of two keys - an "answer" key with the 'yes or no' classification, and an "explain" key that provides a short explanation.

This is an ONNX int4 quantized version of slim-boolean-phi-3, providing a very fast, very small inference implementation, optimized for AI PCs using Intel GPU, CPU and NPU.

Model Description

• Developed by: llmware
• Model type: phi-3
• Parameters: 3.8 billion
• Model Parent: llmware/slim-boolean-phi-3
• Language(s) (NLP): English
• License: Apache 2.0
• Uses: Boolean question answering
• RAG Benchmark Accuracy Score: NA
• Quantization: int4

Model Card Contact

llmware on github

llmware on hf

llmware website