What Arcee AI does
Arcee AI builds efficient open-weight language models and an end-to-end platform for training, merging, deploying, and monitoring small language models (SLMs). Its thesis is 'frontier performance without frontier pricing': domain-adapted SLMs that run cheaply in a customer's VPC, addressing the cost and privacy concerns of large models for regulated industries like legal, healthcare, insurance, and finance. Recent releases include its Trinity family of open-weight models, distributed via Hugging Face and partners such as OpenRouter, plus a chat playground and developer APIs.
Approach and funding
Arcee pioneered model-merging techniques and continues to release new open-weight models rapidly, pairing them with online reinforcement learning for continuous improvement and SLM-powered agentic workflows. Founded in 2023 by Mark McQuade, Jacob Solawetz, and Brian Benedict and based in Miami, Arcee raised a $5.5M seed in early 2024 followed by a $24M Series A led by Emergence Capital. It partners with NVIDIA, Microsoft, AWS, and Hugging Face, positioning itself as a leader in small, efficient, deployable models for enterprises that want control over their AI stack.