Jason AI
Mini toolsTrinity Mini
New
Text
Gen 7
Overview
Trinity Mini is a 26B-parameter (3B active) Trinity MoE model with 128K context, tuned for strong reasoning, function calling and multi-step agents while remaining efficient for enterprise backends.
Description
Trinity Mini is Arcee’s medium-size Trinity model: 26B total parameters with 3B active per token in a sparse MoE setup. It keeps a 128K-token context window and optimized attention to handle long documents, chain-of-thought and multi-tool workflows. Instruction tuning emphasizes reasoning quality, robust function calling and multi-step agent execution, making it a general workhorse for production APIs, on-prem deployments and GGUF-based local serving.
Last updated: January 30, 2026
