🧰 OpenAI API (ML API)

OpenAI API (ML API)#

A hosted API for accessing OpenAI’s models programmatically.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

The OpenAI API is a cloud machine-learning API: it exposes OpenAI’s models — LLMs, embeddings, fine-tuned variants — through simple HTTP endpoints. Rather than training and hosting a model yourself, you send a request, the cloud runs inference, and JSON comes back — inference-as-a-service.

What it offers#

The surface is broad. Chat and text generation (GPT-family models) for Q&A, summarisation and reasoning. An embeddings endpoint turning text into vectors for semantic search, clustering and RAG. A fine-tuning endpoint to specialise a model on your data. Moderation for unsafe content, vision/multimodal for images alongside text, and speech both ways (text-to-speech and Whisper transcription). Plus batch/async processing for volume.

Calling it#

You POST to a REST endpoint (or use an SDK) and read the response:

from openai import OpenAI
client = OpenAI()

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "Explain Bayesian inference simply."}],
)

Benefits and limits#

The appeal is no training or hosting, automatic scale, and simple integration that improves as OpenAI ships new models. The constraints: it is cloud-only (needs internet), adds network latency, costs per token of input and output, and raises privacy questions — sensitive data must clear HIPAA/GDPR review before it leaves your systems.

Theme: ML Platforms & Tools · All terminology

Hint

Mind map — connected ideas

AWS SageMaker · Vertex AI · AWS SageMaker Endpoints · Embedding · Cloud Inference with Big Payloads · Online Experimentation Platforms

Hint

More in ML Platforms & Tools

AWS SageMaker · Google Experiments · Kaggle · ONNX (Open Neural Network Exchange) · TPU Clusters · Vertex AI