One key, one dashboard, every Claude model — ready for Claude Code, Cursor and production.
New accounts get $2.50 in free test usage.
Anthropic-equivalent balance at a discount — spend it only when requests run.
See pricingGenerate a key, point your tool at one endpoint, and every Claude model is available on the same balance.
Generate one sk-* key in your dashboard.
Claude Code, Cursor, Cline, Continue, Blackbox or SDK.
One endpoint gives access to all Claude models.
All Claude tiers, per-key limits and usage in a single dashboard.
Anthropic Messages API and OpenAI-compatible clients out of the box.
SSE streams for coding agents, IDEs and production calls.
Caps, request history and token-level usage clarity.
The most capable Claude model runs on the same key and the same balance — no upgrade, no separate plan.
Explore modelsBuy discounted Anthropic-equivalent credits once, then spend them only when API requests run.
For active teams, heavier coding sessions and production calls.
Buy balanceStreaming and non-streaming calls are billed identically. Every token type is metered and visible in your dashboard.
Everything you send in the prompt and context window.
Every token the model generates in its reply.
Prompt-cache writes, billed once per cached block.
Reused cached context at a reduced rate.
Extended-reasoning tokens, metered separately.
Short answers to the questions that usually matter before buying an API key.
Create a key in minutes and put every Claude model behind one endpoint.