Modern AI APIs go far beyond text generation. Understanding the full spectrum of capabilities helps you build richer applications.

Vision / Image Understanding

Send images to AI models for analysis:

const response = await openai.chat.completions.create({
  model: "gpt-5",
  messages: [{
    role: "user",
    content: [
      { type: "text", text: "Describe this image in detail" },
      { type: "image_url", image_url: { url: "https://example.com/photo.jpg" } }
    ]
  }]
});

Use cases: product image analysis, document OCR, chart interpretation, accessibility descriptions, visual QA.

Image Generation

Generate images from text descriptions:

const image = await openai.images.generate({
  model: "dall-e-3",
  prompt: "A minimalist logo for a tech startup called 'Flux'",
  size: "1024x1024",
  quality: "hd"
});
const imageUrl = image.data[0].url;

Speech-to-Text (Transcription)

Convert audio to text with Whisper:

const transcription = await openai.audio.transcriptions.create({
  model: "whisper-1",
  file: audioFile,
  language: "en"
});

Use cases: meeting transcription, podcast show notes, voice commands, accessibility.

Text-to-Speech

Generate natural-sounding audio from text:

const audio = await openai.audio.speech.create({
  model: "tts-1-hd",
  voice: "nova",
  input: "Welcome to our application!"
});

Use cases: audiobook narration, accessibility, voice assistants, language learning.

Embeddings

Convert text to numerical vectors for search and similarity:

const embedding = await openai.embeddings.create({
  model: "text-embedding-3-small",
  input: "How do I reset my password?"
});
// Returns a 1536-dimensional vector

Use cases: semantic search, recommendations, clustering, RAG (Retrieval-Augmented Generation).

Building a RAG System

RAG combines embeddings with chat completions for knowledge-grounded answers:

1. Index — Embed your documents and store vectors in a database (pgvector, Pinecone)
2. Retrieve — When a user asks a question, find the most relevant document chunks
3. Generate — Pass retrieved context + question to a chat model
4. Cite — Include source references in the response

RAG is the standard pattern for building AI systems that answer questions about your specific data.