One balance for coding model routes

AI token packs for Cursor, OpenCode and agents.

Buy prepaid Quota Tokens and use coding model routes through a clean OpenAI-compatible API. One balance, simple usage logs, model multipliers, and instant setup for your tools.

OpenAI-compatibleCursor-readyOpenCode-readyCard & crypto top-upsUsage logs
Quota balance
138.4M
Live usage
2.1M charged today
Route selector
Choose a coding route
OpenAI-compatible
quota/fast
Quick tasks, low-cost coding
x1
quota/balanced
Daily coding and chat
x2
quota/sonnet
Stronger coding route
x3
quota/sonnet-pro
Advanced reasoning and code edits
x4
API key
qta_live_ab12e8h9k2v3... hidden after creation
Create a key per tool, track usage per key, and revoke access without touching the rest of your setup.
Recent usage
Cursor laptop
quota/sonnet
21,090
OpenCode desktop
quota/balanced
6,000
Agent runner
quota/opus
61,800

How it works

Top up once, issue a key, and route requests from your coding tools.

Quota keeps usage, route selection, and key management in one clean developer surface.

1

Buy Quota Tokens

Choose a prepaid token pack and top up with card, USDC, or USDT.

2

Create your API key

Issue separate keys for Cursor, OpenCode, Cline, and your own agents.

3

Use any supported route

Send standard OpenAI-compatible requests and track route multipliers in one place.

Token packs

Prepaid packs sized for tests, daily coding, and long-running agents.

Different routes consume Quota Tokens at different multipliers. Choose the pack that matches your usage pattern and top up when needed.

Full pricing

Starter

For quick tests and setup

10M
$5 / €5
  • Prepaid Quota Tokens
  • Works across Cursor, OpenCode and agents
  • Usage logs per API key

Builder

Best for daily coding

Most popular
100M
$29 / €29
  • Prepaid Quota Tokens
  • Works across Cursor, OpenCode and agents
  • Usage logs per API key

Pro

For heavy Cursor/OpenCode usage

500M
$99 / €99
  • Prepaid Quota Tokens
  • Works across Cursor, OpenCode and agents
  • Usage logs per API key

Max

For agents and continuous development

Best value
1B
$179 / €179
  • Prepaid Quota Tokens
  • Works across Cursor, OpenCode and agents
  • Usage logs per API key

Route multipliers

Clear route costs before you send traffic.

Each route uses the same API shape, but different route multipliers so you can choose cost and capability deliberately.

RouteMultiplierStatus
quota/fast
Compatibility label: fast
x1
operational
quota/balanced
Compatibility label: balanced
x2
operational
quota/sonnet
Compatibility label: sonnet
x3
operational
quota/sonnet-pro
Compatibility label: sonnet-pro
x4
degraded
quota/opus
Compatibility label: opus
x6
operational
quota/opus-max
Compatibility label: opus-max
x8-x10
operational

Compatibility

Setup guides for the tools developers already use.

Use Quota with Cursor, OpenCode, Cline, or your own OpenAI-compatible clients without changing your core request format.

Cursor

Add Quota as a custom OpenAI-compatible provider and choose your model alias.

OpenCode

Point your provider config to the Quota base URL and keep one shared balance.

Cline

Bring your own route selection, API key, and usage controls for agent workflows.

Custom API

Use curl, JavaScript, Python, or any OpenAI-compatible client with the same base URL.

Dashboard preview

Watch balance, logs, keys, and route usage without guessing.

Quota surfaces recent requests, route selection, charged tokens, and latency so developers can track spend and tune their workflows.

Balance
138.4M
24h requests
158
Error rate
1.8%

Security and clarity

Built for paid usage with clear boundaries.

  • Your own Quota API keys
  • Usage logs per key
  • No raw prompts stored by default
  • Independent service, not affiliated with model providers

FAQ

Practical answers before you set up your key.

How do route multipliers work?

Token packs include base Quota Tokens. Different model routes consume Quota Tokens at different multipliers depending on route cost.

Do I get my own API key?

Yes. Each account can create Quota API keys, rename them, scope routes, and revoke them independently.

Are routes always available?

Route availability can vary. We continuously monitor latency, errors and capacity, and show status publicly.

Do you store prompts?

No raw prompts are stored by default. You can prefer metadata-only retention in settings when available on your account.