Azurecontainerapps

Locate objects in a supplied image with tags, confidence scores, and bounding boxes.

Return a caption, visual tags, and located objects for a supplied image.

Convert raw HTML already held by an agent into deterministic, clean, LLM-ready Markdown without fetching the source URL. Returns the title, byline, excerpt, Markdown, and character count.

Generate one English sentence describing a supplied image.

ManyFish location resolver for forward geocoding. Convert an address, landmark, city, postal code, or place name into coordinates, normalized address components, IANA time zone, and current UTC offset.

Extract printed and handwritten text from a supplied image using OCR.

Convert a document-like image into Markdown while preserving reading order and tables.

ManyFish location resolver for reverse geocoding. Convert latitude and longitude into a normalized address, structured address components, IANA time zone, and current UTC offset.

Extract normalized pages, paragraphs, tables, sections, and geometry from a supplied PDF.

Convert a supplied PDF into Markdown while preserving document structure and tables.

Extract printed and handwritten text from a supplied PDF using OCR.

Fetch a public URL containing static or server-rendered HTML and convert it into clean, LLM-ready Markdown. Does not execute JavaScript. Returns source URLs, article metadata, Markdown, and truncation status.

Search the live web once and return a concise answer with source citations. Supports country context and an optional allowlist of source domains.