Vicuna-13B-16k

Vicuna-13B-v1.5-16k is an advanced chat assistant developed by LMSYS, designed for research in large language models and chatbots. It is fine-tuned from the Llama 2 model using user-shared conversations from ShareGPT.

Key features:
- Auto-regressive language model based on the transformer architecture
- Fine-tuned using supervised instruction fine-tuning and linear RoPE scaling
- Trained on approximately 125,000 conversations collected from ShareGPT.com
- Evaluated using standard benchmarks, human preference, and LLM-as-a-judge methodologies

How it works:
Vicuna-13B-v1.5-16k is an auto-regressive language model that generates text based on the input provided. It leverages the transformer architecture to process and understand the context of the input, allowing it to generate coherent and relevant responses.

Integrations:
- Command line interface: FastChat CLI
- APIs: OpenAI API, Huggingface API

Use of AI:
Vicuna-13B-v1.5-16k is primarily intended for research on large language models and chatbots. It is designed for researchers and hobbyists in the fields of natural language processing, machine learning, and artificial intelligence.

AI foundation model:
Vicuna-13B-v1.5-16k is fine-tuned from the Llama 2 model, which serves as its foundation.

How to access:
- Command line interface: FastChat CLI (https://github.com/lm-sys/FastChatvicuna-weights)
- APIs: OpenAI API, Huggingface API (https://github.com/lm-sys/FastChat/tree/mainapi)

The model is available under the Llama 2 Community License Agreement. Detailed training information and evaluation results can be found in the appendix of the research paper (https://arxiv.org/abs/2306.05685).