Skip to content
OneAPIDocs
ModelsOpen console

Guides

Plans

A plan is a set amount of usage for a fixed period, with higher rate limits while it runs. This page lists what each plan includes, which models it pays for, and roughly how many tokens that usage buys.

Plans at a glance

PlanPriceUsage includedFree web searchRequests per minuteAt onceModels
DeepSix$2/day$351,000 searches, 1,000 page reads200201 of 26
Uncensored Go$5/7 days$10100 searches, 100 page reads3023 of 26
Go$7/month$281,000 searches, 1,000 page reads5056 of 26
Plus$13/month$583,000 searches, 3,000 page reads100106 of 26
Uncensored Pro$15/14 days$35300 searches, 300 page reads100103 of 26
Pro$25/month$10010,000 searches, 10,000 page reads2002017 of 26
Uncensored Max$50/21 days$110600 searches, 600 page reads200203 of 26

Usage included is what the plan can spend over its whole period, at each model's listed price. Some plans also cap a 5-hour or weekly window, which can pause usage before the total runs out; your live windows are on the Subscription page. When a plan has nothing left, requests move to your balance only if extra usage is on. See Which credit pays.

Free web search is a count of web searches and page reads the plan pays for outright each period. They come off before the usage included, never count against the 5-hour or weekly windows, and reset on renewal. Past that count, each one is paid from the usage included at its listed price.

How the token estimate is worked out

A plan's usage is a dollar amount, and every model prices cached input, new input and output differently. So “how many tokens” depends on the mix. The figures below assume 95% cache hits, 3% new input, 2% output, which is close to a coding agent that resends a long, cached conversation on every turn:

tokens = usage ÷ (0.95 × cached input price + 0.03 × input price + 0.02 × output price) × 1,000,000

The pricing page turns that into requests, taking one request as about 40K tokens (a coding agent turn with its cached conversation): requests = tokens ÷ 40,000. “Per 7 days” uses the most the plan's caps allow in a week. A plan's “up to” figure is its usage included times the highest usage multiplier among its models, so it is reached on the boosted model, not on every model.

Estimated tokens by model

ModelBlended / 1MBalance, $10DeepSix, $35Uncensored Go, $10Go, $28Plus, $58Uncensored Pro, $35Pro, $100Uncensored Max, $110
mimo-v2.6-flash$0.01803MNot includedNot included2.25B4.65BNot included8.03BNot included
gpt-6-luna$0.02635MNot includedNot includedNot includedNot includedNot included6.35BNot included
glm-5.3-flash$0.02465MNot includedNot included1.3B2.7BNot included4.65BNot included
minimax-m3.1-flash$0.03325MNot includedNot included911M1.89BNot included3.25BNot included
mimo-v2.6-pro$0.03295MNot includedNot included827M1.71BNot included2.95BNot included
deepseek-v4.1-flashFree$0.03292M1.02BNot included819M1.7BNot included2.92BNot included
gpt-5.6-luna$0.03292MNot includedNot includedNot includedNot includedNot includedNot includedNot included
claude-haiku-5.5$0.08127MNot includedNot includedNot includedNot includedNot included1.27BNot included
deepseek-v4.1-flash-uncensored$0.08120MNot included120MNot includedNot included419MNot included1.32B
glm-5.3-flash-uncensored$0.09113MNot included113MNot includedNot included395MNot included1.24B
gpt-6.1-terra$0.09113MNot includedNot includedNot includedNot includedNot included1.13BNot included
glm-5.3$0.1193.8MNot includedNot includedNot includedNot includedNot included938MNot included
gemini-3.8-flash$0.1284.7MNot includedNot includedNot includedNot includedNot includedNot includedNot included
muse-spark-1.3$0.1375.5MNot includedNot includedNot includedNot includedNot includedNot includedNot included
glm-5.3-uncensored$0.1662.5MNot included62.5MNot includedNot included219MNot included688M
gemini-4-argon$0.1856.3MNot includedNot includedNot includedNot includedNot includedNot includedNot included
claude-sonnet-5.5$0.2540.2MNot includedNot includedNot includedNot includedNot included402MNot included
gpt-6.1-sol$0.2540.2MNot includedNot includedNot includedNot includedNot included402MNot included
gpt-5.6-sol$0.3231.7MNot includedNot includedNot includedNot includedNot includedNot includedNot included
gpt-5.6-terra$0.3429.2MNot includedNot includedNot includedNot includedNot includedNot includedNot included
kimi-k3.1-pro$0.3627.8MNot includedNot includedNot includedNot includedNot included278MNot included
kimi-k3$0.4422.5MNot includedNot includedNot includedNot includedNot included225MNot included
claude-opus-5.5$0.5020.1MNot includedNot includedNot includedNot includedNot included201MNot included
deepseek-v4-pro-0813$0.5418.6MNot includedNot included52.2M108MNot included186MNot included
claude-fable-5.1$1.238.13MNot includedNot includedNot includedNot includedNot included81.3MNot included
gpt-6-astra$1.576.35MNot includedNot includedNot includedNot includedNot included63.5MNot included

“No charge” means the model has no per-token price, so using it never draws the amount down; rate limits still apply. “Not included” means the plan does not pay for that model: calls to it are charged to your balance when extra usage is on, and refused with 403 plan_model_not_allowed when it is off.