Trinity Mini is a 26B-parameter sparse MoE from Arcee AI with 3B active parameters per forward pass. It handles function calling and multi-step agent workflows at low per-token cost, trained end-to-end in the United States.
import { streamText } from 'ai'
const result = streamText({ model: 'arcee-ai/trinity-mini', prompt: 'Why is the sky blue?'})Trinity Mini is a sparse mixture-of-experts model with 26B total parameters and 3B active per forward pass. The compact active footprint keeps inference costs low while the full parameter set provides enough capacity for function calling and multi-step agent workflows.
Arcee AI trained the model end-to-end in the United States, which teams cite for sovereignty and sourcing reviews.
Use https://docs.arcee.ai/language-models/trinity-mini-26b for weights and licensing detail.