ANT for iOS
Build AI-powered iPhone and iPad apps with one API. Chat, vision, OCR, image generation, web search, and function calling — all routed intelligently across 20,000+ models. No model management, no provider SDKs.
Works with any OpenAI-compatible Swift SDK
let config = OpenAI.Configuration(
apiKey: "your-ant-api-key",
baseURL: "https://antbase.ai/v1"
)
let client = OpenAI(configuration: config)
// Chat with intelligent routing
let response = try await client.chat(
model: "ant:auto",
messages: [.user("Analyze this photo")]
)
// OCR — scan a document
let ocr = try await client.chat(
model: "ant:ocr",
messages: [.user([
.image(photoData),
.text("Extract all text")
])]
)Every AI capability for mobile
One API endpoint. No separate SDKs for vision, OCR, or image generation.
Chat Completions
Text generation with streaming. Build chat interfaces, assistants, and conversational features.
Vision & Image Analysis
Send photos from the camera or gallery. ANT routes to the best vision model — Gemini, GPT-4o, or Claude.
OCR / Document Scanning
Extract text from photos via ant:ocr. Scan receipts, labels, documents — get structured markdown back.
Image Generation
Generate images from text prompts via ant:image-gen. Gemini Flash Image, GPT-5 Image, and Flux models.
Web Search
Give your app real-time information with ant:search. The AI can search the web and cite sources.
Function Calling
Define tools and let the AI call them. Build agents that interact with your app's native features.
Recommended models for mobile
| Model | Best For | Typical Speed |
|---|---|---|
| ant:auto | Default — smart routing for everything | Varies |
| ant:fast | Instant responses — autocomplete, quick Q&A | <300ms |
| ant:vision | Camera/photo analysis, image understanding | ~1-2s |
| ant:ocr | Document/label/receipt scanning | ~1s |
| ant:free | Zero cost — great for development/testing | ~500ms |
| ant:image-gen | Generate images from text descriptions | ~3-5s |
Real-world: MyInner.Health
MyInner.Health is an iOS app that uses ANT to power its supplement scanning pipeline. Users photograph a supplement label, and the app runs it through OCR (ant:ocr), LLM parsing with web search (nest:vision + ant:search), and a vision fallback — all routed through a single Antbase API key.
See all use cases →Ship AI on iOS, fast
One SDK. Every AI capability. $5 free credits. No credit card required.