Starter
Monthly plan
$5/month
VANSA / PRICING
For routing, scoring and yes/no decisions.
Pay only for what Vansa reads.
Open sign-up · monthly plans or prepaid credit
$0.034
USD per 1 million input tokens
≈ €0.03 EUR
Prepaid credit: any amount from $1 to $1,000.
Purchased tokens carry over until used.
PLANS
Prices in US dollars. Output tokens are free. All API keys in your account share its token balance.
Monthly plan
$5/month
Monthly plan
$25/month
Monthly plan
$99/month
Any amount from $1 to $1,000
$0.034/1M input tokens
Monthly plans renew automatically on your billing date until you cancel. Each paid billing month includes the plan’s input tokens; unused monthly tokens expire at the end of that billing month and do not roll over. The total, including any applicable tax, is shown before payment. See the terms of service.
New verified accounts can receive a one-time welcome bonus of 10,000,000 tokens. One per person; a bonus can be held for review, for example for duplicate, disposable or automated sign-ups.
YOUR USAGE
Choose a token volume to see its estimated cost. The same rate applies across question types.
For example, 1,000 requests averaging 1,000 input tokens each use 1 million tokens.
≈ €0.03 EUR
1,000,000 input tokens
1,000,000 ÷ 1,000,000 × $0.034
USD is the base rate. EUR is an indicative conversion. This calculator does not create a charge.
COMPARE YOUR USAGE
Estimated total · USD1,000,000 input tokens
1:1 estimate. LLM output tokens equal input tokens. Actual output varies; short classification answers often use fewer tokens.
$0.034
$0.60
$2.80
$6.00
| Model | Input | Output |
|---|---|---|
| Vansa · vansa-3.5 | $0.034 | No charge |
| OpenAI · GPT-6 Luna | $0.10 | $0.50 |
| Google · Gemini 3.5 Flash-Lite | $0.30 | $2.50 |
| Anthropic · Claude Haiku 4.5 | $1.00 | $5.00 |
Standard paid, uncached text prices; batch and cache discounts, free tiers and tool fees are excluded. OpenAI rates apply to prompts up to 272K input tokens per request. Equal input counts are assumed; actual token counts differ by model and workload. This compares prices, not accuracy or speed.
Selected models are available to new customers and have no announced retirement at the date checked. Pricing and availability may change.
WHAT COUNTS
Your context, questions and answer options contribute to the input. Images count as input tokens at the same rate. The API returns the recorded total in usage.input_tokens.
Vansa returns decisions and probabilities. Output tokens are zero, so there is no separate output-token cost.
A successful request deducts its reported input tokens from your account balance. A failed request is not charged; its temporary reservation is released. A request is charged once the model has answered it, even if your client closed the connection before the response arrived, including while the request was still waiting in the queue.
A monthly plan renews automatically and includes its input tokens for each paid billing month; unused monthly tokens do not roll over. Prepaid credit is a one-time purchase from $1 to $1,000 at $0.034 per million input tokens. It does not renew, and purchased tokens carry over until used. The token amount and rate shown for a purchase are fixed for that purchase. Requests use monthly plan tokens first, then promotional tokens such as the welcome bonus, then prepaid credit.
Try a scenario immediately in the public playground. For your own integration, create an account in the console with your email address, Google or GitHub. Then create an API key and choose a monthly plan or prepaid credit.
Yes, in the console. Cancellation takes effect at the end of the paid period; until then, the plan’s tokens remain usable. A plan change takes effect at the next renewal, with no immediate prorated charge.
Requests made with an API key from the console are declined with HTTP 402 and the code insufficient_tokens until you add prepaid credit, choose a plan or your plan renews. Each request reserves enough tokens to cover it before it runs and is then charged only its reported input tokens, so a request can be declined when your balance is nearly used up.
No. Personal API keys issued before the public launch keep working as before and are not billed.
Request-rate limits are assigned individually to each API key and may be unlimited. Size limits are the same for every key: 16,384 tokens per question's prompt and at most 49,152 tokens the model reads per request. See the limits for your key in the console or with GET /v1/me.
The USD estimate is divided by the reference rate of €1 = $1.1403 USD, dated . This is a dated conversion, not a live exchange rate. ECB reference ↗
HOW IT WORKS
Rank search results. Check AI answers. Organize documents. Choose what happens next in an app or a game. Explore the same three steps across different kinds of software.
“We were billed twice for March.”
An excerpt from a customer’s email. Send the full message as context.
“Which team should handle this?”
Define your teams as the possible answers. Vansa scores each option.
Billing99.2% probability
Your helpdesk uses the answer to route the ticket to Billing.
Your context. Your choices. A decision your software can use.
Try this use caseOne email can answer several questions. Switch between the team, urgency, and risk of the customer leaving to see each result.
“Hi, we were billed twice for March. Please refund the duplicate today or we will cancel our plan.”
GET STARTED
Bring your next idea to life with Vansa. Create an account and make your first API call.
Already have an account? Sign in
Sign up with your email, Google or GitHub.
Generate a key in your console. Keep it on your server.
Plans from $5/month or $1 prepaid. No output fees.
Follow the quickstart to connect your application.