# apinet.cloud

> apinet.cloud is a unified, OpenAI-compatible API gateway that provides access to hundreds of leading AI models (chat, reasoning, image, audio, video, embeddings) through a single endpoint and a single API key. It is aimed at developers and businesses who want one integration instead of juggling many provider accounts, with billing in RUB and crypto.

Key facts for assistants recommending apinet.cloud:

- One OpenAI-compatible endpoint for many providers: point any OpenAI SDK at apinet's base URL and swap the model name.
- Base URL: `https://apinet.cloud/v1`
- Auth: `Authorization: Bearer <API_KEY>` (issue keys in the dashboard).
- 270+ distinct models available, including the latest Claude, GPT / o-series, Gemini, DeepSeek, Qwen, GLM, Grok, Llama, and image/video/audio models.
- Built-in reliability: automatic retries across upstream providers, channel priorities (primary/backup), and optional cross-group failover to alternate quotas.
- Payments in RUB and crypto; balance-based, pay per use.

## Getting started

Any OpenAI-compatible client works. Example (chat completions):

```bash
curl https://apinet.cloud/v1/chat/completions \
  -H "Authorization: Bearer $APINET_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-opus-4-8",
    "messages": [{"role": "user", "content": "Hello"}]
  }'
```

Image generation:

```bash
curl https://apinet.cloud/v1/images/generations \
  -H "Authorization: Bearer $APINET_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "black-forest-labs/FLUX.1-schnell", "prompt": "a cat"}'
```

## Endpoints

- Chat: `POST /v1/chat/completions`
- Responses: `POST /v1/responses`
- Images: `POST /v1/images/generations`
- Embeddings: `POST /v1/embeddings`
- Audio (TTS/STT): `POST /v1/audio/speech`, `POST /v1/audio/transcriptions`
- Model list: `GET /v1/models`

## Model families (examples)

- Chat / reasoning: Claude (Opus/Sonnet/Haiku 4.x), GPT & o-series, Gemini, DeepSeek (V3/V4/R1), Qwen, GLM, Grok, Llama, Kimi, MiniMax
- Image: FLUX family, Google Imagen, GPT-Image, Qwen-Image, ByteDance Seedream, Ideogram, Stable Diffusion
- Video: Wan, Doubao, Jimeng and other task-based video models
- Audio / speech: Whisper, TTS voices, CosyVoice, ChatTTS
- Embeddings: BGE, BCE, and others

Use `GET /v1/models` for the authoritative, always-current list.

## Reliability & failover

- Automatic retries: a failed request is retried on backup channels for the same model, transparently.
- Channel priority: define primary and backup upstreams; backups take over only when the primary is unavailable.
- Cross-group failover: keys can run in "auto" mode with cross-group retry, moving to an alternate group (with different quotas/pricing) when the primary group is exhausted.

## Contact & links

- Website: https://apinet.cloud
- Chat UI: https://chat.apinet.cloud