Billing
Tensoras.ai uses a pay-as-you-go USD balance system for inference, embeddings, and reranking. This guide covers pricing, account balance, spending controls, and usage tracking.
How Billing Works
You add funds to your Tensoras account balance (in USD), and usage is deducted from your balance in real time. Every API call is metered by the number of input and output tokens processed, and the per-token cost depends on the model.
There are no minimum commitments. You only pay for what you use.
Inference Pricing
Pricing is listed per million tokens. Input tokens are the tokens in your prompt (system message, user messages, tool definitions, etc.). Output tokens are the tokens the model generates.
| Model | Input (per M tokens) | Output (per M tokens) |
|---|---|---|
llama-3.3-70b | $0.20 | $0.60 |
qwen-3-32b | $0.10 | $0.30 |
deepseek-r1-distill-70b | $0.15 | $0.45 |
codestral-latest | $0.30 | $0.90 |
llama-3.1-8b | $0.05 | $0.10 |
mistral-7b-instruct | $0.04 | $0.08 |
Example Cost Calculation
A chat completion request using llama-3.3-70b with 1,000 input tokens and 500 output tokens:
Input: 1,000 tokens x ($0.20 / 1,000,000) = $0.0002
Output: 500 tokens x ($0.60 / 1,000,000) = $0.0003
Total: $0.0005This $0.0005 is deducted directly from your USD balance.
Embedding Pricing
| Model | Price (per M tokens) |
|---|---|
bge-large-en-v1.5 | $0.01 |
Embedding costs are based on input tokens only. There are no output tokens for embedding requests.
Reranking Pricing
| Model | Price (per M tokens) |
|---|---|
bge-reranker-v2-m3 | $0.02 |
Reranking costs are based on the total tokens across the query and all candidate documents in a single rerank request.
Adding Funds
Top-Up Packages
Larger top-up amounts include volume bonuses:
| Package | You Pay | Balance Added | Bonus |
|---|---|---|---|
| $10 | $10 | $10.00 | — |
| $50 | $50 | $52.50 | 5% |
| $100 | $100 | $110.00 | 10% |
| $200 | $200 | $230.00 | 15% |
Via the Console
- Go to cloud.tensoras.ai and navigate to Console > Billing.
- Click Add Funds or scroll to the Top Up Balance section.
- Select a package and confirm payment.
Funds are added to your balance immediately and do not expire.
Via the API
from tensoras import Tensoras
client = Tensoras()
# Add $50 in funds (amount in cents)
client.billing.funds.add(amount=5000)import Tensoras from "tensoras";
const client = new Tensoras();
// Add $50 in funds (amount in cents)
await client.billing.funds.add({ amount: 5000 });Spending Limits
Set a daily spending limit to prevent unexpected overages. When the limit is reached, API requests will return a 402 Payment Required error until the next day (UTC midnight) or until the limit is raised.
Set a Limit via Console
- Navigate to Console > Billing > Spending Limits.
- Enter your daily limit (e.g., $10/day).
- Click Save.
Set a Limit via API
client.billing.spending_limits.update(daily_limit=1000) # $10.00 in centsawait client.billing.spending_limits.update({ dailyLimit: 1000 }); // $10.00 in centsTip: Start with a conservative daily limit and increase it as you understand your usage patterns. This prevents a misconfigured loop from draining your account balance.
Usage Tracking
The Tensoras Console provides detailed breakdowns of your API usage.
Console Dashboard
Navigate to Console > Usage to view:
- Per-model breakdown — see token usage and cost for each model
- Per-key breakdown — identify which API keys are driving usage
- Daily and monthly trends — track usage over time with graphs
- Per-request logs — inspect individual requests with token counts and latency
Usage API
Query your usage programmatically:
usage = client.billing.usage.retrieve(
start_date="2026-02-01",
end_date="2026-02-17",
)
for entry in usage.daily:
print(f"{entry.date}: {entry.total_tokens:,} tokens, ${entry.cost:.4f}")const usage = await client.billing.usage.retrieve({
startDate: "2026-02-01",
endDate: "2026-02-17",
});
for (const entry of usage.daily) {
console.log(`${entry.date}: ${entry.totalTokens.toLocaleString()} tokens, $${entry.cost.toFixed(4)}`);
}Plan Pricing
In addition to pay-as-you-go inference costs, Tensoras offers plan tiers that determine your rate limits, Knowledge Base allowance, and storage:
| Plan | Monthly Cost | RPM | Knowledge Bases | Storage |
|---|---|---|---|---|
| Developer | $29 | 600 | 5 | 5 GB |
| Pro | $49 | 3,000 | 10 | 10 GB |
| Enterprise | Custom | 10,000 | Unlimited | Custom |
Plan fees are billed monthly and are separate from pay-as-you-go inference costs. You can upgrade or downgrade your plan at any time in Console > Billing.
Invoices and Payment Methods
- Payment methods: Credit card and ACH (US bank transfer). Enterprise customers can pay via invoice.
- Invoices: Monthly invoices are generated on the first of each month and available in Console > Billing > Invoices.
- Receipts: Automatic email receipts are sent for every top-up purchase and plan payment.
Next Steps
- Rate Limits — understand and handle rate limits
- Authentication — manage API keys
- Quickstart — make your first API call