Pricing
Pricing
Pay as you go at per-token prices, or a monthly Agent plan that covers your usage up to a 5-hour and a weekly limit — on the same key and the same endpoints.
Agent plan
The Agent plan is a monthly subscription that pays for your requests — API and Playground, every model, the same API key and endpoints — up to a 5-hour and a weekly usage limit. Usage is counted at the prices you would pay from your balance, so the limits compare directly with the price list. It renews every month until you cancel; payments are non-refundable except where required by law, and a cancelled plan keeps working until the end of the period you paid for. Upgrade at any time — you pay only the prorated difference; a downgrade starts with your next period.
Starter
Pro
Team
How the limits work
The 5-hour window starts with your first request on the plan; when it ends, your next request starts a new one. The weekly window runs in whole weeks from the day your plan started. When either limit is reached, your requests continue on your balance if "At a limit, continue on my balance" is on (it is off by default); otherwise the API answers 402 with a Retry-After header until the next reset. A key set to Balance only on the API keys page never uses the plan, and a failed request counts nothing.
Any API call, not only coding
The plan is not only for coding tools. Coding agents, a monitoring loop that sends images every few seconds, a nightly pipeline — every request on the same key and the same endpoints counts, on every model.
Upgrade or downgrade
Upgrade at any time: the new tier starts at once, Stripe charges only the prorated difference for the rest of the billing period, and the new limits apply immediately. A downgrade takes effect at the end of the current period, and you keep your current tier until then.
Pay as you go
Add funds to a balance and pay only for the tokens you use. No minimum term.
Price list
| Model | Input / 1M tokens | Output / 1M tokens | Cached input / 1M tokens |
|---|---|---|---|
| GLM 5.3 Flash10% offuntil Oct 13, 2026, 6:51 PM EDT | |||
| DeepSeek V4.1 Flash10% offuntil Oct 13, 2026, 6:50 PM EDT | |||
| Nemotron 3.5 Lightning 30B | US$0.07 / M Tokens | US$0.17 / M Tokens | US$0.02 / M Tokens |
| Ornith 1.5 35B | US$0.14 / M Tokens | US$0.65 / M Tokens | US$0.025 / M Tokens |
| Qwen3.8 27B | US$0.14 / M Tokens | US$0.65 / M Tokens | US$0.025 / M Tokens |
Accounts that allow all model suppliers in their data-handling settings are billed 5% less than the prices above.
How billing works
Every request is metered in tokens. Input tokens are what you send, including the conversation history and any files converted to text or images. Output tokens are what the model writes, including reasoning tokens when a model thinks before answering. When part of your input was already processed recently, those cached input tokens are billed at the lower cached price where a model lists one. A request that fails is not charged.
An example
A request with 10,000 input tokens and 1,000 output tokens on GLM 5.3 Flash costs US$0.0022.
Paying
Your balance is held in US dollars. You can pay in USD or CAD; a CAD payment is converted to a USD credit at a rate locked at checkout. Payments are processed by Stripe.
Welcome credit
New accounts that verify their email currently receive a welcome credit of US$1.60.