Billing & Wallet
How Tolka Edge billing works, including wallet top-ups, GPU-time metering, and balance protection.
How Tolka Edge billing works, including wallet top-ups, GPU-time metering, and balance protection.
Tolka Edge uses a prepaid wallet for dedicated GPU usage. You add credits to your wallet, and your balance is charged based on the time your AI Rig remains active.
The wallet model
Your wallet holds credits that are used to pay for Tolka Edge services.
For dedicated AI Rigs:
- Billing is based on GPU time, not tokens.
- The applicable GPU rate is shown before deployment.
- Billing starts when your Rig begins booting.
- Billing continues while the GPU remains allocated to your Rig.
- Usage is tracked to the second.
- Billing stops when the Rig is terminated.
For example, if a GPU costs ₹63/hour and your Rig runs for 10 minutes, you are charged approximately ₹10.50. You are not charged based on the number of input or output tokens generated.
Top up your wallet
Add credits from the Billing Dashboard. After a successful top-up, the new balance becomes available for deployments and usage. Make sure your wallet has enough credits before starting a dedicated Rig.
UPI payments
Tolka Edge supports wallet top-ups through UPI. Complete your top-up from the Billing Dashboard and your wallet balance will update after the payment is confirmed.
Balance protection
Before starting a paid operation, Tolka checks that your account has sufficient wallet balance.
For an active dedicated Rig, Tolka continuously monitors the remaining balance. When your balance becomes too low to safely continue running the Rig, Tolka may stop or terminate the Rig to prevent uncontrolled spending.
Dedicated GPU billing
Each deployed Rig has a fixed GPU rate for its active session. For example:
GPU rate: ₹63/hour
Runtime: 12 minutes 30 seconds
Charge: ₹63 × (12.5 / 60) = ₹13.13The actual charge is based on the precise active runtime recorded for the Rig.
No token-based billing
Tolka Edge's dedicated GPU offering is not priced per token. Your inference volume can vary without changing the GPU rate:
- 100 tokens → same GPU rate
- 10,000 tokens → same GPU rate
- 1,000,000 tokens → same GPU rate
As long as the Rig remains allocated, billing is based on elapsed GPU time.
Viewing your usage
You can view your balance, spending, and dedicated GPU usage from the dashboard.
The Dedicated GPU Dashboard shows the current Rig state and active runtime, while the Billing Dashboard shows wallet transactions and charges.