zephyr-mistral-7b-chat-ov

zephyr-mistral-7b-chat-ov is an OpenVino int4 quantized version of Zephyr-Mistral-7B-Chat, providing a fast inference implementation, optimized for AI PCs using Intel GPU, CPU and NPU.

zephyr-mistral-7b-chat is a leading and very popular chat fine-tune of Mistral.

Model Description

• Developed by: Huggingface
• Quantized by: llmware
• Model type: Mistral
• Parameters: 7 billion
• Model Parent: HuggingFaceH4/zephyr-7b-beta
• Language(s) (NLP): English
• License: Apache 2.0
• Uses: General purpose chat
• RAG Benchmark Accuracy Score: NA
• Quantization: int4

Model Card Contact

llmware on github

llmware on hf

llmware website