Engram and Harvey trained a legal agent that cut query costs and improved accuracy

Engram · 19 August 2026, 09:30

Engram and Harvey trained a legal AI agent that combines parametric knowledge, text notes, and search to handle law firm queries. The agent achieved $0.13 per query at 30% accuracy versus $1.32 per query at 25% for Claude Opus 4.8 on a synthetic 100M-token legal firm dataset. The trained agent learns to perform targeted searches instead of exhaustive document reads, reducing inference costs by an order of magnitude while improving accuracy.

Why it matters

A legal agent trained on a 100M-token mock law firm reduced average query cost from $1.32 to $0.13 compared to Opus 4.8 while raising all-pass accuracy from 25% to 30%.