Understand this story
Hightouch tested scoring for over-merge detection in identity resolution (IDR) using the decision model Jev and GPT-6 Luna. The tests found that Jev and GPT-6 Luna used alone achieved 88.8%–97.3% accuracy, with the decision model costing $0.5 to $0.10 per 1,000 identities and responding in about 0.15 seconds, but on harder benchmarks it was about 6 percentage points lower than Claude Opus. Jev's confidence calibration is better, with most errors accompanied by low confidence, so it performed better than Luna in a hybrid setup; combining Jev with Claude Opus into a hybrid review setup can reduce costs by more than 4x while approaching Opus-level accuracy.
AI-generated from public reports
Latest development · Report
Hightouch: Using Jev + Claude Opus for identity resolution review, cutting costs by 4xDevelopments
1 confirmed developments
Hightouch tests Jev and Luna for identity resolution scoring
Source:English text is being prepared.
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 4
Hacker News Trending
Hightouch: Using Jev + Claude Opus for identity resolution review, cutting costs by 4x
Hightouch testing found that combining the decision model Jev with Claude Opus into a hybrid review approach can approach Opus's accuracy on identity resolution (IDR) over-merge detection while cutting costs by more than 4x. Used alone, Jev and GPT-6 Luna achieved accuracy of 88.8%–97.3%, with the decision model costing $0.5–$0.10 per 1,000 identities and responding in about 0.15 seconds, but on harder benchmarks it was about 6 percentage points lower than Opus. Jev's confidence calibration is better, with most errors accompanied by low confidence, so it performs better than Luna in the hybrid approach.