Model Information

Meta-Llama 3-8B-Instruct

Questro uses the Meta-Llama 3-8B-Instruct model from Hugging Face. It is an instruction-tuned large language model with 3 billion parameters.

Hugging Face Link: https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct

Key Features:

• Instruction-tuned: The model is trained on a dataset of instructions and responses, making it better at following instructions and generating text that aligns with the user’s requests.

• Generative: The model can generate various creative text formats, including stories, articles, poems, code, scripts, musical pieces, email, letters, etc.

• Contextual Understanding: The model has been trained on a vast dataset, giving it a good understanding of language and the ability to provide contextually relevant answers.

Note: The model is available for free use and is subject to the Hugging Face Terms of Service.