Unified AI gateway · Usage-based pricing

One API for AI models, routing and multimodal workloads

Build with GPT, Claude, Gemini, Grok, DeepSeek and multimodal model families through one managed gateway. ModAPI adds unified endpoints, routing, metering, key management and support around every request.

One managed key for hundreds of public model routes
OpenAI-compatible, Claude-style and Gemini-style workflows
Usage logs, model-level pricing and multimodal routing
Start Building Explore Models
200 ok
POST /v1beta/models/{model}:generateContent
Request
curl -X POST "/v1beta/models/{model}:generateContent" \
-H "x-goog-api-key: sk-••••" \
-d '{
"contents": [
{ "role": "user",
"parts": [{ "text": "..." }] }
]
}'
Response
{
"candidates": [{ "content": { "parts": [{ "text": "Gemini request served." }] } }],
"usageMetadata": { "totalTokenCount": 25 }
}
93ms 25tokens illustrative example
usage metered
400+public model routes
1managed API key
USDusage-based billing
24/7usage visibility

Core Capabilities

One platform for modern AI APIs

AI development no longer depends on a single model or a single provider. ModAPI gives teams fast switching, broad model access, predictable integration patterns and better cost control.

Unified metering

Model-level prices with clear billing units

Review the applicable input, output, cache, request, image or duration unit before choosing a model. The published ModAPI rate is the basis for credit deduction.

Unified access

One API key for hundreds of AI models

Access GPT, Claude, Gemini, Grok, DeepSeek and multimodal model families without maintaining a separate integration for every provider.

Familiar endpoints

Use the API style that fits your stack

Keep OpenAI-compatible, Claude-style and Gemini-style workflows instead of rewriting every application around one schema.

Multimodal catalog

Explore models across providers and modalities

Browse chat, reasoning, image, audio, video, embedding, rerank and realtime endpoints before choosing the right model.

LLM router

Route requests across models and providers

Switch models, test alternatives and keep applications connected when a first-choice model is too expensive, slow or unavailable.

Built for developers and AI product teams

ModAPI gives teams a smaller integration surface for working with many AI models, while keeping usage, access and multimodal workflows manageable.

Usage tracking

Monitor costs and activity from one gateway.

Team controls

Manage keys, access and account workflows.

Pricing & Credits

USD pricing you can review before use

The Model Catalog shows the applicable billing unit for each route, including input, output and cache prices per 1M tokens where relevant, plus request, image or duration units for supported media models.

Pay and account in USD

Credit purchases, published prices and usage deductions are denominated in United States dollars.

Digital delivery

After a successful payment, purchased credits are normally added immediately. Payment-review or system exceptions are handled within one business day.

Usage-based deduction

Credits are deducted using the selected model's published units and the actual metered API usage or completed task.

Service credits, not money

Credits can only be used for ModAPI API services. They cannot be transferred, withdrawn, redeemed for cash or used as a payment instrument.

Review Model Pricing Refund Policy Delivery Policy

How It Works

From model choice to production routing

1

Choose

Browse providers, modalities and pricing patterns before selecting the right model for your product.

2

Connect

Use one API key with OpenAI-compatible, Claude-style and Gemini-style endpoint workflows.

3

Route

Route requests, track metered usage and keep applications connected while retaining one operational view across model families.