Deci introduces DeciLM 6B and DeciLM 6B-Instruct, new Large Language Models (LLMs) that offer high performance and computational efficiency. With 5.7 billion parameters, these models establish a new benchmark for inference efficiency and speed, and can be used in a wide range of generative AI applications.
DeciLM's unique architecture was generated using AutoNAC, Deci's Neural Architecture Search engine. Coupled with Deci's inference SDK, these models offer substantial throughput enhancement, resulting in improved user experience, reduction in inference cost, and a smaller carbon footprint.