lmsys/vicuna-7b-delta-v0

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Apr 6, 2023Architecture:Transformer0.2K Cold

The lmsys/vicuna-7b-delta-v0 is a 7 billion parameter delta model developed by LMSYS, fine-tuned from the LLaMA architecture. This model is specifically designed as a chat assistant, trained on user-shared conversations from ShareGPT. It serves primarily as a research tool for large language models and chatbots, requiring application on original LLaMA weights to become a functional Vicuna model.

Loading preview...

Vicuna-7b-delta-v0 Overview

This model is a 7 billion parameter delta model developed by LMSYS, fine-tuned from the original LLaMA architecture. It is specifically designed as a chat assistant, leveraging supervised instruction fine-tuning on approximately 70,000 conversations collected from ShareGPT.com. It's important to note that this is a delta model and cannot be used directly; it must be applied on top of the original LLaMA weights to form a complete Vicuna model. Instructions for this process are available in the FastChat repository.

Key Capabilities

  • Chat Assistant: Primarily functions as a conversational AI.
  • Research Tool: Intended for research and development in large language models and chatbots.
  • LLaMA-based: Built upon the robust LLaMA transformer architecture.

Good for

  • Researchers and Hobbyists: Ideal for those studying natural language processing, machine learning, and artificial intelligence.
  • Chatbot Development: Provides a foundation for building and experimenting with conversational agents.
  • Understanding Fine-tuning: Offers a practical example of instruction fine-tuning on real-world conversational data.