Pricing

See what a turn costs before you send it.

Credits are fixed per model and call type. Pick a plan by its monthly credits and the turns they cover, then add model access when you need it.

Plan options

Free

For trying Qriple and lightweight daily use.

$0/month
  • Up to 2 models per turn
  • Trial: 3 turns (6 trial credits) at signup. The trial does not reset.
  • 0 monthly credits after the trial
  • 10 modelsFree includes GPT-6 Luna, Claude Haiku 5.5, Gemini 3.8 Flash, Gemini 3.5 Flash Lite, Gemini Flash Lite (latest), Grok 4.3, Free Models Router, Nemotron 3.5 Lightning, Liquid LFM 2.5, Gemma 4 31B
  • Chat and review history
Start free

Lite

For regular cross-checks with the core paid models.

$19.99/month
  • Up to 2 models per turn
  • 1,480 credits per month, about 493 turns with Sonnet 5.5 as main and GPT-6.1 Terra as reviewer
  • +5 modelsEverything in Free, plus GPT-6.1 Terra, Claude Sonnet 5.5, Gemini 3.1 Pro, Gemini Pro (latest), Grok 4.7
  • Native web search and image inputs
  • Email support
Upgrade to Lite
Recommended

Pro

For steady research and work where a wrong answer costs something.

$59.99/month
  • Up to 3 models per turn
  • 4,520 credits per month, about 155 turns with Opus 5.5 as main and GPT-6.1 Sol and Gemini 3.1 Pro as reviewers
  • +2 modelsEverything in Lite, plus GPT-6.1 Sol and Claude Opus 5.5
  • Native web search and image inputs
  • Email support
Upgrade to Pro

Max

For broader model coverage and the most monthly credits.

$149.99/month
  • Up to 3 models per turn
  • 11,360 credits per month, about 315 turns with Opus 5.5 as main and GPT-6 Astra and GPT-6.1 Sol as reviewers
  • +2 modelsEverything in Pro, plus GPT-6 Astra and Dots3-Note Preview
  • Same advanced inputs as Pro
  • Email support
Start with Max

Plan matrix

Features by plan

Model access, trial and monthly credits, and advanced inputs for each plan.

Features by plan table
FeatureFreeLiteProMax
Model access
10 modelsFree includes GPT-6 Luna, Claude Haiku 5.5, Gemini 3.8 Flash, Gemini 3.5 Flash Lite, Gemini Flash Lite (latest), Grok 4.3, Free Models Router, Nemotron 3.5 Lightning, Liquid LFM 2.5, Gemma 4 31B
+5 modelsEverything in Free, plus GPT-6.1 Terra, Claude Sonnet 5.5, Gemini 3.1 Pro, Gemini Pro (latest), Grok 4.7
+2 modelsEverything in Lite, plus GPT-6.1 Sol and Claude Opus 5.5
+2 modelsEverything in Pro, plus GPT-6 Astra and Dots3-Note Preview
Models per turnUp to 2 modelsUp to 2 modelsUp to 3 modelsUp to 3 models
Monthly credits0 monthly credits. Trial: 3 turns (6 trial credits)1,480 credits (about 493 turns with Sonnet 5.5 as main and GPT-6.1 Terra as reviewer)4,520 credits (about 155 turns with Opus 5.5 as main and GPT-6.1 Sol and Gemini 3.1 Pro as reviewers)11,360 credits (about 315 turns with Opus 5.5 as main and GPT-6 Astra and GPT-6.1 Sol as reviewers)
Advanced inputsText-first core workflowWeb search + image inputsSame as LiteSame as Pro
SupportCommunity supportEmail supportEmail supportEmail support

How credits work

What each model costs, by call

Credits are fixed per model and call type, so you can see what a call costs before you send it.

Credit cost per model and role
ModelMain answerReviewerMain answer, deep mode
Claude Sonnet 5.52218
GPT-6.1 Terra115
GPT-6.1 Sol448
Gemini 3.1 Pro10931
Claude Opus 5.5161552
GPT-6 Astra181659

credits per call

FAQ

Frequently asked questions

Need more detail before choosing a plan? Start with these answers.

Yes. You can move between Free, Lite, Pro, and Max as your usage changes.

The trial at signup is 3 turns, which is 6 trial credits on eligible models. It does not reset. Free has no monthly credits after the trial. Paid plans get a monthly credit allowance, and when it is used up, new turns are paused until the next reset date shown on your billing page.

A credit is our usage unit. One credit is about $0.005 of estimated model inference cost. Each model has a fixed credit cost for a main answer and for deep mode, so you can see what a call costs before you send it.

Stronger models use more compute for each answer, and some of them reason internally before they reply. Those reasoning tokens are billed as output, so a stronger model uses more credits for the same answer length.

No. A reviewer writes a short review of about 200 visible tokens rather than a full answer, so it costs far fewer credits. If you switch to a reviewer's full answer, that switch is charged as a main answer for the model you switched to.

Yes. Deep mode lets a main answer think longer, so it costs more credits. Depending on the model, a deep main answer costs about 2 to 9 times the default. Deep mode is not available on Free.

Yes. Marking where the answers disagree uses 0 credits. Only model answers are charged.

If you cancel before the model sends any output, the call costs 0 credits. If output has started, the full call is charged. Credits are not partially refunded.

Free includes GPT-6 Luna, Claude Haiku 5.5, Gemini 3.8 Flash, Gemini 3.5 Flash Lite, Gemini Flash Lite (latest), Grok 4.3, Free Models Router, Nemotron 3.5 Lightning, Liquid LFM 2.5, and Gemma 4 31B. Lite adds GPT-6.1 Terra, Claude Sonnet 5.5, Gemini 3.1 Pro, Gemini Pro (latest), and Grok 4.7. Pro adds GPT-6.1 Sol and Claude Opus 5.5. Max adds GPT-6 Astra and Dots3-Note Preview.

No. You can start on the pricing plans without entering a personal API key. If we ever require one for a specific workflow, we will make that explicit before checkout.

See /refund-policy for the current refund, renewal, and cancellation guidance. The checkout flow will show the checkout provider, purchase terms, and any applicable tax details before you pay, and support can help with account-specific questions.