Deci AI's profile picture

Deci AI


Research interests

NAS, Foundation models, Inference efficiency.

Organization Card
About org cards

About Us

Deci is on a mission to empower AI teams with advanced tools to remove development barriers and attain production-grade performance and efficient inference faster.

If you have questions, please feel free to reach out to us on Twitter, our website or join our Deep Learning Daily Discord community.

DeciLM 6B:

A 5.7 billion parameter, permissively licensed, pretrained LLM using variable Grouped-Query Attention (GQA) to achieve an optimal balance between performance and computational efficiency. Generated using Deci's proprietary Neural Architecture Search technology, AutoNAC, DeciLM 6B delivers 15 times the throughput of Llama 2 7B while maintaining comparable quality. DeciLM 6B and its fine-tuned variant, DeciLM 6B-Instruct, consistently rank among the top-performing open-source LLMs in the 7 billion parameter category across various LLM evaluation tasks.


An open-source, permissively licensed text-to-image latent diffusion model generated by Deci's Neural Architecture Search technology.
DeciDiffusion delivers a 3x increase in throughput while generation Stable Diffusion caliber images.


An open-source, permissively licensed 1B parameter code generation model generated by Deci's Neural Architecture Search technology.
Equipped with a 2048-context window, DeciCoder delivers a x3.5 increase in throughput, improved accuracy on the HumanEval benchmark, and smaller memory usage compared to widely-used code generation LLMs such as SantaCoder.


The most advanced inference SDK for LLM optimization and deployment, Infery-LLM includes unique features such as optimized kernels, continuous batching, advanced selective quantization, ultra-efficient beam search, parallel execution, and more. Book a personalized demo to explore Infery-LLM's capabilities.


Open-0source Object Detection foundational model generated by Deci's Neural Architecture Search technology. YOLO-NAS is a game-changer in the world of object detection, providing superior real-time object detection capabilities and production-ready performance.


An open-source library for training PyTorch-based computer vision models. Developed by Deci's deep learning experts for the benefit of the AI community. Use SuperGradients to boost performance with advanced techniques. Easily train or fine-tune SOTA computer vision models for all the main tasks with one training library.


DataGradients is an open-source Python-based library specifically designed for computer vision dataset analysis. It automatically extracts features from your datasets and combines them all into a single user-friendly report.

Deci Deep Learning Platform:

Simplify and accelerate the development of computer vision and NLP and Generative AI applications with highly accurate and efficient foundation models and advanced tools to customize, optimize, and deploy them to production.

Deci is powered by groundbreaking Automated Neural Architecture Construction (AutoNAC™) technology. Deci's AutoNAC™ engine democratizes the use of Neural Architecture Search for every organization and helps teams quickly generate fast, accurate, and efficient deep learning models.


None public yet