Weave QuickDistill

Quick Start:

1. Sync traces: Fetch a Weave project to import traces from your LLM calls
2. Select Evaluation data: Choose a subset of traces to use as input for weak model evaluation and click 'Export selected to test set'
3. Generate weak outputs: Run inference with smaller models on your test set
4. Evaluate quality: Use judges to compare weak model responses against strong model outputs to find the best budget model

Run Weak Model Inference

Run Evaluation

Settings

Test Judge

Test your judge on sample data to see exactly what inputs/outputs it receives

⚡ Run End-to-End Test

This will automatically: Export selected traces → Run weak models → Evaluate with judge

Running End-to-End Test