Changelog

Initial Release

• Phase 1: Multi-provider GPU price comparison (Vast.ai, RunPod, Lambda) with deferred-provider tracking for TensorDock
• Phase 1: One-click instance provisioning with SSH access
• Phase 1: Real-time pricing dashboard with filters and sorting
• Phase 2: Autopilot mode — automatic cost optimization
• Phase 2: Auto-migration when savings exceed 15% threshold
• Phase 2: Spot instance recovery in under 30 seconds
• Phase 2: Budget enforcement with daily caps and alerts
• Phase 3: Scheduled GPU jobs with cron-based timing
• Phase 3: Webhook notifications for instance events
• Phase 3: Analytics dashboard with spending insights
• Phase 3: Template library for common ML workloads