Live AN1 Pilot Demo

This is a live production proof clip showing health check, inference execution, request logging, and measured cost savings against a baseline model. No mock responses.

What you're seeing

  • Health check returns ready: true
  • Inference call returns request_id, latency_ms, and savings_usd
  • Same request_id appears in Partner Portal logs
  • Vercel production logs confirm successful API execution

System validation

Health endpoint: Confirms AN1-Turbo API is operational with model loaded and ready to serve inference requests.

Inference request: Real 256-dimensional z-vector input processed by production model, returning logits, probabilities, and predicted class.

Cost calculation: Transparent baseline comparison showing exact token usage, baseline cost, AN1 cost, and realized savings.

Audit trail: Full request lifecycle captured in Partner Portal logs with timestamps, request IDs, and performance metrics.

Ready to try it yourself?

Contact us at partner@animacore.ai to get your pilot API key and start measuring real savings on your inference workloads.

Built with v0