Run Advanced Reasoning on DigitalOcean with Arcee AI's Trinity Large-Thinking

Run Advanced Reasoning on DigitalOcean with Arcee AI's Trinity Large-Thinking Why this model, why now Built for real-world agent workloads on DigitalOcean A new phase of AI infrastructure Get started in seconds About the author Try DigitalOcean for free Related Articles The Agentic Era Demands a New Class of Infrastructure: DigitalOcean Acquires Katanemo Labs Now Available: DigitalOcean Cloud Security Posture Management (CSPM) NVIDIA GTC 2026 Confirmed It: The Inference Era Is Here By DigitalOcean Published: April 1, 2026 3 min read Today, we’re announcing that Arcee AI ’s Trinity Large-Thinking is now available in Public Preview on DigitalOcean’s Agentic Inference Cloud, giving developers the ability to run frontier-class reasoning workloads without managing infrastructure or stitching together complex systems. DigitalOcean is proud to partner with Arcee to bring Trinity Large-Thinking to AI builders, available via Serverless Inference, on day one. Instantly available and queried directly through the DigitalOcean Cloud Console or API alongside the compute, data, and services you already run on DigitalOcean. Trinity Large-Thinking didn’t emerge in a vacuum. It’s been pressure-tested in exactly the kind of workloads DigitalOcean is built for. Arcee is a 26-person San Francisco startup that spent nine months building a full open-weight model family from the ground up, with the explicit goal of producing models developers and enterprises could actually own. The result is a family ranging from 4. 5B to 400B parameters, and a top-of-stack reasoning model that has earned its place in production. In its first two months, Trinity served over 3.4 trillion tokens on OpenRouter, becoming the most-used open weight model in the U. S. , driven by always-on, agentic workloads running continuously. Trinity Large-Thinking builds on that foundation with extended reasoning, stronger multi-turn tool use, and more stable long-running behavior.