Pricing

Per-token rates

USD per 1M tokens.

Model	Input	Output
flash	$1.25	$2.50
base	$1.25	$3.75
large-v2	$3.75	$6.25
large	$3.75	$6.25

Reasoning tokens count as output. Cached input is billed at the input rate. GET /v1/models returns the live rates.

Web search costs $0.008 per search on top of tokens. See web search.

How billing works

Usage is charged in USD against one prepaid balance shared by every key and model. Add to it by sending USDC on Solana to your deposit address on the dashboard’s Top up page.

Before a request runs we hold enough balance to cover its input and its maximum output, then charge the actual usage when it finishes and release the rest. If your balance cannot cover even a short reply, the request returns 402 insufficient_balance. If it covers less than your max_tokens, the output limit is lowered to what the balance can pay for.

Plans

A monthly plan buys credit at a bonus and raises your rate limits. Plan credit is spent at the rates above, up to a limit per 5-hour session and per week. A session or week starts with the first request it pays for. Past a limit, requests are paid from your balance until it resets; with no balance they return 429 plan_limit_reached with resets_at.

Plan	Price / month	Credit / month	Credit / 5-hour session	Credit / week	Requests / min	Concurrent requests
Pay as you go	—	top-ups only	—	—	60	5
Starter	$20	$23	$1.20	$5.40	60	5
Pro	$100	$120	$6	$28	300	10
Max	$200	$250	$12.50	$58	600	16

Manage your plan from the dashboard.