Arcee AI: Trinity Large Thinking

Arcee AI: Trinity Large Thinking is a text-in, text-out model from Arcee AI with a 262,144-token context window and a maximum completion length of 80,000 tokens. It supports tool use and reasoning, making it suitable for agentic workflows and multi-step problem solving. Structured output support is unconfirmed, so developers who depend on guaranteed JSON or schema-constrained responses should verify that before committing. At $0.25 per million input tokens and $0.80 per million output tokens, Trinity Large Thinking sits in the budget tier relative to many reasoning-capable models. Its blended benchmark score of 39.8 is drawn from a single benchmark, so the performance picture is thin and should be treated with caution rather than as a settled verdict. Teams running long-context reasoning tasks on a tight budget may find it worth a closer look, but anyone making a quality-critical choice should test it directly against their own workloads before relying on that single score.