Blog
Notes from the
model server.
Tutorials, model deep-dives, and engineering notes on running AI workloads at production scale.
Coming soon
We're getting the first set of posts ready. Here's what's on deck:
- Comparison
Claude vs GPT vs Gemini: Which AI API should you pick in 2026?
An honest, hands-on comparison of the three frontier model families across reasoning, code, vision, and cost.
- Engineering
Cut your AI bill by 50% with smart model routing
How to route easy queries to small models and hard queries to flagships, automatically.
- Tutorial
Use ModelServer with Cursor in 2 minutes
Step-by-step guide to point Cursor at any model — Claude, GPT, or Gemini — through ModelServer.
- Story
Why we built ModelServer: one key, every model
The story behind ModelServer and why API fragmentation is the biggest hidden cost in AI engineering.