OpenAI Integration
Every OpenAI model through one SDK face
Every OpenAI model available through a single unified interface. Streaming, function calling, vision, embeddings — plus automatic fallback to Claude or Gemini on rate limits.
Models available
Every model. One SDK.
GPT-4o
Fastest multimodal model — text, vision, and function calling.
GPT-4
Flagship reasoning model for complex tasks.
o1
Chain-of-thought reasoning for math, code, and science.
DALL-E 3
State-of-the-art image generation from text prompts.
Whisper
Production-grade speech-to-text in 100+ languages.
Embeddings
text-embedding-3 for semantic search and RAG pipelines.
Code
One call. Any model.
// Generate with GPT-4o
const result = await ee.ai.generate({
model: 'gpt-4o',
prompt: 'Summarise Q3 revenue trends',
context: await ee.storage.get('reports/q3.pdf'),
})
// Stream a response
const stream = await ee.ai.generate({
model: 'gpt-4o',
prompt: 'Write release notes for v3.12',
stream: true,
})Features
Full OpenAI surface. Unified API.
Streaming
Token-by-token streaming for real-time UIs and chat interfaces.
Function Calling
Structured tool use — let the model call your functions.
Vision
Analyse images, charts, screenshots — multimodal out of the box.
Embeddings
Generate and store vectors for semantic search and RAG.
Automatic fallback.
Hit an OpenAI rate limit? EEv3 auto-switches to Claude or Gemini transparently — no downtime, no code changes.
Connect OpenAI