slim-xsum-phi-3-ov

slim-xsum-phi-3-ov is a specialized function calling model that generates an "extreme summary" (e.g. 'xsum') or headline from a context passage.

This is an OpenVino int4 quantized version of slim-xsum-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-xsum-phi-3
• Language(s) (NLP): English
• License: Apache 2.0
• Uses: Summarization from complex business documents
• RAG Benchmark Accuracy Score: NA
• Quantization: int4

Model Card Contact

llmware on github

llmware on hf

llmware website