01
Top up your balance
Start with a minimum $10 top-up. Your first successful top-up adds bonus credit equal to 50% of the amount actually paid.
$10 minimum · First top-up +50%
Pay for what you use.
From your first experiment to production at scale, use the models you need through one API. Top up from $10, pay for actual usage, and grow without a fixed monthly fee.
No fixed monthly fee$10 minimum top-up50% extra credit on your first successful top-upRates based on model and usage
How billing works
One account balance across models. Each call is measured by the model’s billing unit and shown in your account.
02
Choose a model
Compare model rates before you call. Charges depend on the model, selected specification, billing unit and actual usage.
Tokens · Images · Video time
03
Review your usage
Check request records and costs by model in your account. Use the usage view to follow how your balance is spent.
Dashboard → Usage
| Provider | Model | Category | Price | Details |
|---|---|---|---|---|
| Amazon |  Nova 2 Lite | Text | InputUSD 0.24 / 1M tokensOutputUSD 2 / 1M tokens | View model |
| DeepSeek |  DeepSeek Chat | Text | InputUSD 0.1176 / 1M tokensOutputUSD 0.236 / 1M tokens | View model |
| DeepSeek | DeepSeek Reasoner | Text | InputUSD 0.1176 / 1M tokensOutputUSD 0.236 / 1M tokens | View model |
| DeepSeek | DeepSeek V3.2 | Text | InputUSD 0.16704 / 1M tokensOutputUSD 0.24768 / 1M tokens | View model |
| Arcee Ai |  Trinity Mini | Text | InputUSD 0.036 / 1M tokensOutputUSD 0.12 / 1M tokens | View model |
| OpenAI |  GPT-Image-1.5 | Image | InputUSD 4 / 1M tokensOutputUSD 25.6 / 1M tokens | View model |
| Anthropic |  Claude Opus 4.5 (latest) | Text | InputUSD 4 / 1M tokensOutputUSD 20 / 1M tokens | View model |
| Kling AI |  Kling 2.5 Turbo Pro Image to Video | Video | Output · Up to 2500 characters per request · Generation mode pro · Service route fal · Video duration 5 secondsUSD 0.056 / second | View model |
| Kling AI | Kling 2.5 Turbo Pro Text to Video | Video | Output · Up to 2500 characters per request · Generation mode pro · Service route fal · Video duration 5 secondsUSD 0.056 / second | View model |
 Nano Banana Pro Preview | Image | InputUSD 0.8 / 1M tokensOutputUSD 4.8 / 1M tokens | View model | |
| xAI |  Grok 4.1 Fast | Text | InputUSD 0.16 / 1M tokensOutputUSD 0.4 / 1M tokens | View model |
| xAI | Grok 4.1 Fast (Reasoning) | Text | Input · Context ≤ 1000000 tokens · Maximum output 1000000 tokensUSD 0.16 / 1M tokensOutput · Context ≤ 1000000 tokens · Maximum output 1000000 tokensUSD 0.4 / 1M tokens | View model |
| Gemini 3 Pro Preview | Text | Input · Base context tier (boundary 200000 tokens)USD 1.6 / 1M tokensOutput · Base context tier (boundary 200000 tokens)USD 9.6 / 1M tokens | View model | |
| OpenAI | GPT-5.1 | Text | Input · Service tier flexUSD 0.5 / 1M tokensOutput · Service tier flexUSD 4 / 1M tokens | View model |
| OpenAI | GPT-5.1 Chat | Text | InputUSD 1 / 1M tokensOutputUSD 8 / 1M tokens | View model |
| OpenAI | GPT-5.1 Codex | Text | InputUSD 1 / 1M tokensOutputUSD 8 / 1M tokens | View model |
| OpenAI | GPT-5.1 Codex Max | Text | InputUSD 1 / 1M tokensOutputUSD 8 / 1M tokens | View model |
| OpenAI | GPT-5.1 Codex mini | Text | InputUSD 0.2 / 1M tokensOutputUSD 1.6 / 1M tokens | View model |
| Moonshot AI |  Kimi K2 Thinking | Text | InputUSD 0.48 / 1M tokensOutputUSD 2 / 1M tokens | View model |
| Anthropic | Claude Opus 4.5 | Text | InputUSD 4 / 1M tokensOutputUSD 20 / 1M tokens | View model |
| OpenAI | GPT OSS Safeguard 20B | Text | InputUSD 0.064 / 1M tokensOutputUSD 0.192 / 1M tokens | View model |
| OpenAI | GPT OSS Safeguard 120B | Text | InputUSD 0.12 / 1M tokensOutputUSD 0.48 / 1M tokens | View model |
| Nvidia |  Nemotron Nano 12B v2 VL | Text | Input · Context ≤ 128000 tokens · Maximum output 16384 tokensUSD 0.16 / 1M tokensOutput · Context ≤ 128000 tokens · Maximum output 16384 tokensUSD 0.48 / 1M tokens | View model |
| MiniMax (minimax.io) |  MiniMax-M2 | Text | InputUSD 0.204 / 1M tokensOutputUSD 0.816 / 1M tokens | View model |