Weni/WeniGPT-Mistral-7B-instructBase
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kLicense:apache-2.0Architecture:Transformer Open Weights Cold

WeniGPT-Mistral-7B-instructBase is a fine-tuned variant of the Mistral-7B-Instruct-v0.1 model, developed by Weni. This instruction-tuned model leverages the Mistral architecture, known for its efficiency and strong performance in its size class. It was trained with specific hyperparameters including a learning rate of 0.0004 and 8000 training steps, making it suitable for general instruction-following tasks.

Loading preview...