Chronos 70B V2

About

The Chronos 70B V2 is a sophisticated large language model built on the Llama 2 Base architecture, specifically crafted for activities such as chat, roleplay, and storywriting. It excels in reasoning and logic, producing lengthy, coherent outputs due to extensive training with human inputs. With a context length support of up to 4096 tokens, the model necessitates quantization for use on consumer GPUs, available in formats like GPTQ, AWQ, and GGUF, each offering different performance-resource trade-offs. Utilizing Alpaca formatting for improved outputs, it is offered under a non-commercial Creative Commons license.

Capabilities