AI & Compute

Visual caption generator: turns any image URL into a written description covering the scene, objects, people, and readable text within it. No images are stored during or after processing. A drop-in way for text-based agents and pipelines to gain image understanding for cataloging, moderation pre-checks, or accessibility workflows.

Image-to-text description generator: submit any image URL and receive a detailed written account of what's shown, subjects, setting, actions, and any visible text, rendered entirely in natural language. Processed in memory with nothing retained. Useful for building searchable captions, screening uploads before publishing, or giving text-only agents a way to understand visual content.

UGC product video prompt: the compiled prompt and render plan for a 5s 9:16 energetic spot on seedance-1.5-pro

Text narration API: turns up to 2000 characters of written text into spoken audio, choosing from five male and five female voice options, returned as base64-encoded WAV. Nothing typed is stored. Good for audiobook snippets, accessibility read-aloud features, and adding a voice layer to otherwise silent notification systems.

Voice synthesis API: converts written text (up to 2000 characters) into a natural-sounding spoken-word audio file, delivered as base64 WAV with a choice of five male and five female voice presets. No account or third-party key required, and text is never retained. Suited for voice assistants, IVR prompts, and turning written notifications into audio.

Speech audio generator: reads any block of text aloud and returns the result as a downloadable WAV file, selectable across ten built-in voice presets. Pay-per-call with no subscription or stored text. Useful for narrating short-form content, generating spoken alerts, or producing sample audio for voice-agent testing.

Generates unambiguous alphanumeric secrets, hex strings, and UUIDs from OS-grade entropy in a single call, with length and batch size configurable. For provisioning temporary credentials, signing keys, and identifiers in automated pipelines that need real randomness rather than a plausible-looking string guessed by a model.

Secure token generator API: cryptographically random UUIDs (batch up to 20), hex tokens, or unambiguous alphanumeric secrets (length 8-128) from OS-grade entropy. For request ids, nonces, temporary passwords, and API-key material. Generated locally, never logged.

Automatic image tagging API: generates a ranked list of descriptive keywords for any image URL, spanning what's depicted, its visual style, and dominant colors. Processing happens without storing the source image. Suited for digital-asset-management systems, e-commerce catalogs, and content classification pipelines that need labels without manual review.

Alt-text generator API: WCAG-friendly, under-125-character accessibility alt text for any image URL, generated on our own hardware. Use for automated accessibility compliance, CMS pipelines, and site remediation agents. Images are never stored.

Transparent background generator: takes any image URL, segments out the foreground subject, and returns it as a base64 PNG with full alpha transparency. Nothing from the source image is retained past the request. Useful for sticker packs, thumbnail production, marketplace listings, and any pipeline that needs a subject on its own without a backdrop.

Turn any link or short block of text, up to 2000 characters, into a scannable QR code PNG, delivered as base64 with adjustable pixel size and built-in error correction. Built for onboarding flows, Wi-Fi credential sharing, event check-in badges, and print materials where a URL needs to become something a phone camera can read instantly.

Generates a ready-to-embed QR code image, base64 PNG, from any text or URL input up to 2000 characters, with configurable dimensions and error-correction level. A drop-in way for ticketing systems, packaging, and signage pipelines to produce scannable codes without installing a QR library or maintaining rendering code.

Image alt-attribute generator: takes any image URL and returns a short, descriptive alt-text string sized for the HTML alt attribute, tuned for screen-reader clarity rather than verbose captioning. No image data is kept. Useful for CMS plugins, static-site build steps, and agents auditing pages for missing or poor alt text.

Accessibility alt-text generator: produces a concise, screen-reader-ready description under 125 characters for any image URL, matching WCAG guidance on alternative text. Images pass through without being retained. Built for content-management pipelines, bulk site remediation, and automated accessibility audits that need alt attributes filled in at scale.

Cryptographically secure randomness for agents that are deterministic or sandboxed. Two modes: raw bytes (?bytes=1..1024, default 32, clamped) returning hex and base64url; or uniform integers (?min= and ?max= together, optional ?count=1..1000) rejection-sampled so there is no modulo bias. Errors: 400 bad_bytes for a non-integer, bad_range when max is not greater than min, range_too_large. Generated server-side, and the response says so - if nobody may have seen it, generate it yourself.

Translate supplied text through MyMemory with source-language autodetection or an explicit language code.

One featured space image or video per day, with a written explanation, media type, and direct image URL, as structured JSON. Optional date param retrieves any past day's picture. Built for content-generation agents, daily-newsletter bots, and novelty/education feeds that want a fresh space visual with zero curation effort.

Semantic vector generation: turns batches of text (up to 64 per request) into dense 768-dimension embeddings ready to feed into a vector database or similarity search index. Runs without any external API key or subscription, pay only for the call. Well suited for building RAG retrieval layers, clustering related content, or scoring how similar two passages are.

Determines the language of any input text with a confidence score and runner-up candidates, useful when a short snippet is ambiguous between closely related languages. For content routing, translation triage, and multilingual pipelines that need a fast, deterministic language call instead of an LLM guess.

QR code generator API: turn any text or URL (up to 2000 characters) into a scannable PNG QR code, returned as base64 — size configurable 128-1024px, medium error correction. Generated locally in milliseconds; input never stored. Use for payment links, event tickets, Wi-Fi sharing, onboarding flows, and print materials.

Translates text between about 100 languages with Meta M2M100 1.2B: pass the text and the target language (ISO code like es, fr, de, zh, ja, or its English name); the source language is detected if omitted. Long texts are split by sentence, so paragraphs keep their order and line breaks. Up to 2000 characters per call. Pairs with speech-to-text: translate a transcript, then read it aloud. You are only charged if the translation is delivered. No API key, no account.

Render text (URL, vCard, Wi-Fi config, …) as a QR code — SVG string or PNG data URL.

Edits an image you provide instead of generating one from scratch: pass the image and a black-and-white mask, and Stable Diffusion inpainting repaints only the white area following your prompt. Image and mask go as https URLs or as base64, up to 10 MB each. Returns a hosted image URL valid for 24 hours. You are only charged if the edit is delivered. No API key, no account.