An AI API puts a hosted model behind an HTTP call — classification, summarisation, sentiment, moderation, extraction — so your code gets the answer without you renting GPUs, picking weights, or keeping an inference server alive. One key covers every model here, billed per call.
Also called an LLM API, machine learning API, inference API or model API55 services in this categorySend text and optional language to detect profanity in user-generated content. Useful for moderation checks before publishing comments, chats, or reviews.
External ProvidersAnalyze a YouTube video URL and get its theme plus extracted quotes. Useful for summarising content, indexing videos, and surfacing key lines.
Detect the language of a text sample and get language, ISO codes, script, and confidence. Useful for routing multilingual content and NLP workflows.
Send text and get entity-level sentiment results in a Google-backed response. Useful for reviews, social posts, and support messages.
Summarize PDF, DOCX, DOC, or ODT files from an upload or URL. Choose short, medium, or long output and get back data.summary.
Classify text with Google-based analysis. Send text and receive a structured result under data.google for routing, moderation, or text analysis.
Summarize plain text or a web page using text, url, summary_length, and output_language inputs. Useful for article digests, research, and content triage.
Send text and a requested service to get moderation results back in a data.google object. Useful for filtering user-generated content before publishing.
Extract document data from a URL or uploaded file. Returns Azure or ApyHub output for invoices, forms, reports, and other documents.
Transcribe WAV audio from a URL or uploaded file using a language locale like en-US. Returns the transcript as plain text in data.
Extract text from a video URL or uploaded file using Azure or Google OCR. Get provider response data for captions, labels, and on-screen text.
Detect shot boundaries in uploaded videos or URLs using Azure or Google. Returns service-specific results for scene segmentation and video analysis.
The question you arrived with, and the endpoint that answers it.