How Decagon Runs 90% of Its Agents on Open-Source Models

Why it matters

Decagon, an AI company specializing in customer support automation, runs 90% of its agents on open-source models. In this interview, the team shares why they chose the open-source path, how they achieve high reliability through model routing and fine-tuning, and the practical benefits in cost, data privacy, and iteration speed — offering valuable insights for AI deployment.

Key takeaways

• Open-source models offer significant cost and control advantages for large-scale deployment, especially in high-concurrency scenarios.
• Dynamic model routing switches between open-source and proprietary models based on task complexity.
• Fine-tuning and RAG enable open-source models to approach the performance of proprietary counterparts.
• The fast-moving open-source ecosystem reduces reliance on single API vendors and improves system resilience.