Vigostral-7B-Chat: A French Conversational LLM

Vigostral-7B-Chat is a 7 billion parameter large language model, part of the Vigogne LLMs family, developed by bofenghuang. It is fine-tuned from the Mistral-7B-v0.1 base model, specifically optimized for French language chat applications. The model utilizes a prompt template adapted from Llama-2's chat format, ensuring consistent and effective conversational interactions.

Key Capabilities & Features

• French Language Optimization: Primarily designed and fine-tuned for generating responses in French.
• Conversational AI: Excels in chat-based interactions, following a structured prompt template.
• Mistral-7B-v0.1 Base: Built upon the robust Mistral-7B-v0.1 architecture.
• Quantized Versions Available: Readily available in AWQ, GPTQ, and GGUF formats via TheBloke for efficient inference on various hardware.
• Flexible Inference: Supports inference using Hugging Face Transformers and vLLM, including an OpenAI-compatible API server.

Usage Considerations

• License: A significant portion of its training data is distilled from GPT-3.5-Turbo and GPT-4; users should exercise caution to comply with OpenAI's terms of use.
• Limitations: Still under development, it may occasionally generate harmful, biased, incorrect, or unhelpful content.