Guides
How to choose a model, verify your usage and read your bill, with checks you can run yourself.
2026-10-26 · 6 min read
What is a Claude Code relay? How it works and how to set one up
Where a relay sits between Claude Code and Anthropic, how it differs from going direct, and every step from install to your first request.
Read the guide →
2026-10-26 · 4 min read
Opus 5.5, Fable 5.1, Sonnet 5.5 or Haiku 5.5: prices and roles in Claude Code
The four models side by side at list price, and how to split work across the main session, subagents and light tasks.
Read the guide →
2026-10-26 · 6 min read
How to choose a Claude Code relay: checks you can run yourself
Pricing, sourcing, billing transparency, client restrictions, and checks you can run with a small balance.
Read the guide →
2026-10-26 · 5 min read
How to verify the model and usage of every request
Confirm the model in Claude Code, check token usage in the console log, and recompute the charge at list price.
Read the guide →
2026-10-26 · 5 min read
Cache hit rate and billing explained
How cache writes and reads are priced, how to read your hit rate, and why long sessions get cheaper.
Read the guide →