Nolano AI is a startup focused on building efficient on-prem deployment of foundation models.
Pitch Deck: Link | Twitter: Link
Our Progress So Far:
- Developed a proprietary Turbo LLM Inference Engine: Link
- Introduced LoRD, a novel paradigm for large language model compression: Paper Link
- Released an open-source library for running LLMs locally: Link
- Testing the initial version with a few Clients
- Released Hi-NOLIN, the best performing 9B Hindi-English Bilingual LLM: Link
Founders:
Tejas Vaidhya: Co-founder/ Chief Executive Officer, Previously at Mila, CERN, ETH Zurichs, IIT Kharagpur.
Ayush Kaushal: Co-founder/ Chief Technical Officer, Previously at Research at Google, UT Austin, IBM, ETH Zurich, IIT Kharagpur.
Irina Rish: Co-founder/ Chief Research Officer, Full professor at Universite de Montreal and Mila. Head of Autonomous AI Lab. Canada Excellence Research Chair, CIFAR Chair, INCITE project PI
More info:
Nolano's Blogs: Link
Research: Link
Paper on LoRD (A novel compression paradigm we developed): Link
Inference Engine Benchmarks: Link