# Apertis AI — Full Service Description

> One API for 500+ AI models. Coding Plan subscriptions for Claude Code, Cursor, Aider, and OpenClaw.

## About

Apertis AI is a unified API gateway that provides developers and teams access to 500+ AI models from 30+ providers through a single OpenAI-compatible API. Drop-in replacement for OpenAI SDK — switch models with one line of code.

## API Endpoints

All endpoints are OpenAI-compatible and require an API key (Bearer token).

### Text Generation
- `POST /v1/chat/completions` — Chat completions (GPT-5.4, Claude Opus 4.6, Gemini 3.1 Pro, etc.)
- `POST /v1/completions` — Legacy text completions
- `POST /v1/messages` — Native Anthropic Messages API format

### Other Modalities
- `POST /v1/embeddings` — Text embeddings
- `POST /v1/images/generations` — Image generation (DALL-E, Flux, etc.)
- `POST /v1/audio/speech` — Text-to-speech
- `POST /v1/audio/transcriptions` — Speech-to-text (Whisper)
- `POST /v1/audio/translations` — Audio translation

### Models
- `GET /v1/models` — List all available models
- `GET /v1/models/:model` — Get model details

### Base URL
```
https://api.apertis.ai/v1
```

## Supported Providers

OpenAI, Anthropic, Google (Gemini), AWS Bedrock, Azure OpenAI, Meta (Llama), Mistral, Alibaba (Qwen), Baidu, Tencent, Zhipu, Cohere, DeepSeek, xAI (Grok), Coze, Cloudflare Workers AI, Ollama, and more.

## Featured Models

### Frontier Models
- GPT-5.4, GPT-5.4 Codex
- Claude Opus 4.6, Claude Sonnet 4.6
- Gemini 3.1 Pro, Gemini 3 Flash
- Kimi K2.5
- MiniMax M2.1

### Free Models (No Cost)
- DeepSeek V3.2
- Gemini 3 Flash
- GPT-5.1 Codex Mini
- Llama 4 Scout
- Qwen 3 235B
- And more

## Pricing

### Pay-as-you-go
No monthly fee. Pay only for what you use with competitive per-token pricing.

### Coding Plans (Monthly Subscriptions)
- **Lite** — $12/mo: For individual developers
- **Pro** — $25/mo: Higher quotas for power users
- **Max** — $200/mo: Maximum throughput for teams

All plans include free prompt cache, auto-failover, and access to 500+ models.

## Key Features

### Drop-in OpenAI Replacement
```python
from openai import OpenAI
client = OpenAI(
    api_key="your-apertis-key",
    base_url="https://api.apertis.ai/v1"
)
response = client.chat.completions.create(
    model="claude-sonnet-4-5-20250514",
    messages=[{"role": "user", "content": "Hello!"}]
)
```

### Web Search Integration
Add `:web` suffix to any model ID for real-time web search:
```python
response = client.chat.completions.create(
    model="gpt-5.2:web",
    messages=[{"role": "user", "content": "Latest AI news today"}]
)
```

### Free Prompt Cache
Automatic prompt caching reduces costs on repeated prefixes. No configuration needed.

### Auto-Failover
When a provider experiences downtime, requests automatically route to backup channels with compatible models.

## SDK & Tool Integrations

- **Python SDK**: OpenAI Python SDK (drop-in compatible)
- **Vercel AI SDK**: Official provider
- **LiteLLM**: Supported proxy target
- **LlamaIndex**: Official integration
- **Kilo Code**: Official provider

### Compatible Coding Tools
- Claude Code
- Cursor
- OpenClaw
- Aider
- Cline
- Continue

## Links

- Website: https://apertis.ai
- API Documentation: https://docs.apertis.ai
- Model Catalog: https://apertis.ai/models
- Pricing & Subscribe: https://apertis.ai/subscribe
- Enterprise: https://apertis.ai/enterprise
- Status Page: https://status.apertis.ai
- Changelog: https://apertis.ai/changelog
- GitHub: https://github.com/apertis-ai
- Compare vs OpenRouter: https://apertis.ai/compare/openrouter
- Compare vs Together AI: https://apertis.ai/compare/together-ai
- Compare vs LiteLLM: https://apertis.ai/compare/litellm
- Terms of Service: https://apertis.ai/terms
- Privacy Policy: https://apertis.ai/privacy
- Contact: hi@apertis.ai