metadata
license: apache-2.0
inference: false
base_model: llmware/slim-tags
base_model_relation: quantized
tags:
- green
- p1
- llmware-fx
- ov
slim-tags-ov
slim-tags-ov is a specialized function calling model that generates a list of tags, e.g., 'meaningful objects', from a text passage, which is useful for summarization and various retrieval strategies.
This is an OpenVino int4 quantized version of slim-tags, 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: tinyllama
- Parameters: 1.1 billion
- Model Parent: llmware/slim-tags
- Language(s) (NLP): English
- License: Apache 2.0
- Uses: Tag generation, summarization and search/retrieval enrichment
- RAG Benchmark Accuracy Score: NA
- Quantization: int4