Overview

Why Use the Weco Dashboard?

The Weco Dashboard provides a centralized view of all your optimization runs with powerful features:

Visual Progress Tracking: Monitor optimization progress in real-time with interactive charts and metrics
Solution Tree Explorer: Navigate through every code variant generated during optimization to understand the evolution of your solutions
Run Management: Stop and delete runs remotely without the need to access the CLI
Historical Analysis: Compare performance across different runs and time periods
Credit Management: Track your credit usage, view billing history, and manage payments
Cost Transparency: See exactly how many credits each run and optimization step costs
Auto Top-up: Set up automatic credit purchases to avoid interruptions
Persistent Storage: Keep your optimization history safe and accessible from any device

Creating a Weco account ensures your runs are saved and accessible across devices, rather than being limited to local CLI history.

weco run --source optimize.py \
     --eval-command "python evaluate.py" \
     --metric accuracy --goal maximize

For an example, check out the Quickstart guide.

Visit https://dashboard.weco.ai and sign in with the same Google account used in the CLI.

Track progress, metrics and solutions across runs with real-time charts.
Explore the solution tree to see every code variant and their performance.
Monitor credit usage with detailed cost breakdowns per run and optimization step.
Manage billing through the credits dashboard with usage history and payment options.
Stop and delete runs remotely to control costs and manage resources.
Set up auto top-up to ensure uninterrupted optimization sessions.

Explore these real optimization runs to see the Weco Dashboard in action:

VLM Chart-to-CSV Accuracy - A vision-language model optimization that improved accuracy from 45.33% to 60.02% (+32.4% improvement) over 50 experiments.
Structural-break AUC - An optimization run that improved ROC-AUC score from 0.556 to 0.7805 (+40.4% improvement) over 120 experiments.

These shared runs demonstrate the solution tree, metric progress tracking, and how Weco iteratively improves code performance.