Cline · Roo · Aider · Cursor · caching on

Cheaper Claude for your coding agent.

Coding agents resend the same system prompt and file context every turn. HalfRouter passes Claude prompt caching through natively, so those tokens come back as cheap cache reads — and you pay in USDT, no card.

Free trial credits on signup · pay-as-you-go in USDT
For builders

Agent loops, without the bill shock

Caching that sticks

Long system prompts and file context bill as ~0.1× cache reads instead of full-price input, turn after turn.

</>

Works in your tools

Drop the base URL into Cline, Roo, Aider, Continue or Cursor and pick a Claude model. Done.

Cheap or rock-solid

Standard tier for the lowest price, Dedicated for pure first-party upstreams with cascading failover.

Cline / Roo — API Provider
API Provider  :  OpenAI Compatible
Base URL      :  https://halfrouter.vip/v1
API Key       :  YOUR_HALFROUTER_KEY
Model ID      :  claude-sonnet-4.6
FAQ

Coding on Claude, answered

How does caching save money on coding agents?

Agents resend a large, mostly-identical context every turn. With native prompt caching, the repeated tokens are billed as cheap cache reads instead of full-price input — the savings compound over a session.

Which tools does it work with?

Anything OpenAI-compatible: Cline, Roo Code, Aider, Continue.dev, Cursor, LibreChat, plus the OpenAI SDKs.

Which Claude model should I use for coding?

Sonnet 4.6 is a strong default for agentic coding; Opus 4.8 for the hardest tasks; Haiku for cheap, fast edits.

How do I pay?

Top up in USDT — pay-as-you-go, no card and no KYC.

Your base_url is one line away.

Sign up, grab a key, top up in USDT — your existing OpenAI-style code runs on GPT, Claude & Gemini at half the bill.