Show HN: Echo – Fable-level results at 1/3 the cost using open-weight models

Hacker News featured Echo, an open-source system that combines multiple lightweight models to match Fable-level results at roughly one-third the inference cost. Rather than using a single model for all tasks, Echo dynamically allocates computation and decides which models should participate in each problem—one finding is that individually weaker models can still be highly valuable on specific problems or when part of a combination. Early testing shows Echo reaching Fable-level performance while substantially reducing inference costs, though evaluation on coding and agentic tasks is still underway.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton