> For the complete documentation index, see [llms.txt](https://anyint.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://anyint.gitbook.io/docs/faq.md).

# FAQ

## What is AnyInt?

AnyInt is an AI API aggregation platform. It gives you one gateway for multiple model families, media workflows, subscription controls, and team-level governance instead of asking you to integrate every provider independently.

## Which API styles do you support today?

The currently published catalog includes:

* OpenAI-compatible chat completions
* Anthropic-compatible messages and token counting
* Gemini-compatible content generation and streaming
* DashScope image and video generation routes
* AI Music routes under `/suno/*`

Start with [API Overview](/docs/api-reference/overview.md) if you want the high-level map.

## How do I find valid model IDs?

Use `GET /v1/models` or the model picker in your dashboard. Do not assume a model is enabled for your account until you confirm it there.

Model availability can vary by:

* subscription type
* custom model purchases
* key policy
* organization policy

## Can I use existing SDKs?

Yes. The fastest path is the OpenAI-compatible route with an OpenAI SDK pointed at `https://gateway.api.anyint.ai/openai/v1`.

Use provider-native routes when:

* you need Anthropic-style content blocks or token counting
* you need Gemini-native `generateContent`, streaming, or function declarations
* you want media or AI Music workflows that do not fit a chat-completions client

## Why not just call providers directly?

Teams usually adopt AnyInt because they want:

* one account and key lifecycle
* one model-discovery flow
* one billing and subscription surface
* one place to apply team policy, limits, or routing

## <sup>How does pricing work?</sup>

AnyInt supports one prepaid billing mode and two subscription types:

* Pay As You Go: top up prepaid balance and let API requests deduct from it
* Plan subscriptions: fixed tiers with shared limits across included models
* Custom subscriptions: choose models and limits individually

The two subscription types are mutually exclusive for one active account at a time. Pay As You Go is the separate prepaid billing path for self-serve usage.

## What happens when I hit a limit?

The product docs currently describe:

* daily plan call limits can trigger throttling behavior
* monthly token or generation caps can block further usage until reset or renewal
* expired subscriptions invalidate the associated key
* exhausted prepaid balance can block further usage until you recharge

For production systems, treat subscription state and limit state as part of your operational monitoring.

## Does AnyInt support image and video workflows?

Yes. The currently published catalog includes:

* Gemini image generation
* DashScope image generation
* DashScope video generation with task polling

See [Multimodal Overview](/docs/models-and-modalities/overview.md) and [Media APIs](/docs/api-reference/media-apis.md).

## Does AnyInt support AI Music?

Yes. AI Music is live and available through `/suno/*` routes. The published flows include:

* generation
* continuation
* cover creation
* personas
* lyrics
* stem splitting
* WAV export
* MV generation
* task polling
* webhooks

Start with [AI Music Overview](/docs/ai-music/overview.md).

## Are team and enterprise features available?

Yes, but rollout can vary by customer and environment. The current product plan covers:

* organizations and projects
* team API keys
* key-level permissions
* smart routing policies
* SSO and session security
* enterprise billing and invoice workflows
* event webhooks and notifications

If you publish these pages publicly, some of them may still make sense as hidden or staged content until rollout is complete.

## How should I handle async tasks?

Media and AI Music workflows are not simple request-response APIs. The usual pattern is:

1. create a task
2. store the returned task ID
3. poll the query endpoint or wait for a webhook
4. treat the asset as ready only after final success

## Where should I look for exact request schemas?

Use this GitBook for product-level guidance, then use the published API catalog in Apifox for field-level schema details when wiring production clients.

## Where do I go for support or status?

Use [Support](/docs/support.md) as the central place for support, sales, and status links when you publish the docs.