Keynote-Technology/KAI-7B-v0.1
TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Nov 7, 2023License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

KAI-7B-v0.1 by Keynote-Technology is a 7 billion parameter generative text model, fine-tuned from Mistral 7B, with a 4096 token context length. This model demonstrates strong performance in STEM benchmarks, notably outperforming Meta-Llama 2 70B in tested areas. It is designed as a pretrained base model, excelling in general reasoning but requiring further development in mathematical and coding applications.

Loading preview...