Proxis is the first dedicated platform for LLM distillation and serving, enabling production-ready models at 1/10th the cost. Distill Llama 405b on your own data for 5x faster, 10x cheaper AI than closed-source alternatives.
Proxis lets you distill large language models (LLMs) into smaller, faster models without sacrificing performance. Built for high performance, Proxis focuses on ease of use.
Model developers, AI researchers, startups and companies looking to deploy LLMs efficiently.