bofenghuang/vigogne-13b-chat
Vigogne-13B-Chat is a 13 billion parameter LLaMA-based model developed by bofenghuang, specifically fine-tuned for multi-turn dialogue in French. This model excels at engaging in conversational exchanges as an AI assistant, making it particularly suitable for French-language chat applications. It is designed for research use, adhering to the licensing terms of Stanford Alpaca.
Loading preview...
Vigogne-13B-Chat: A French Chat Model
Vigogne-13B-Chat is a 13 billion parameter LLaMA model developed by bofenghuang, specifically fine-tuned for multi-turn dialogues in French. Its primary purpose is to function as an AI assistant capable of engaging in conversational exchanges with human users in French.
Key Capabilities
- French Dialogue: Optimized for understanding and generating natural, multi-turn conversations in French.
- LLaMA Architecture: Built upon the LLaMA-13B base model, leveraging its robust language understanding capabilities.
- Research-Oriented: Intended and licensed for research use only, aligning with the Stanford Alpaca project's terms.
Usage and Limitations
This model can be easily integrated using the Hugging Face transformers library, with provided Python code examples for inference. A Google Colab notebook is also available for quick experimentation. It's important to note that Vigogne-13B-Chat is still under development and may produce harmful, biased, incorrect, or unhelpful content. Users should be aware of these limitations, especially given its research-only license (CC BY NC 4.0 for the dataset).