Every agent-payable API.

Get ready-made video recipes with hook, shot list, interview questions, equipment, recording time, platforms and follow-up formats (one shoot, several posts). Filter by category, words or exact recipe id. Content in German or English (language: de or en); the response also lists all recipe ids.

Takes a video (up to 24 MB) and one to five SRT or WebVTT subtitle texts with language codes, and returns the video with each as a separate, switchable subtitle track tagged with its language, the first marked default. Picture and sound are copied, not re-encoded, so quality is unchanged and it is fast. MP4 uses mov_text (plays in QuickTime, browsers with players, most phones); MKV uses SRT tracks (VLC, mpv, Plex). Unlike /v1/burn-subtitles nothing is drawn into the picture.

Takes a video (up to 24 MB) and returns the best poster frames: candidates are spread over the running time, and for each segment ffmpeg's thumbnail filter selects the frame closest to the segment average, which avoids transitions, flashes and motion blur. Candidates that are almost black or very flat are dropped. The best candidates come back as JPEG at the chosen width, each with timestamp and mean brightness, ranked by brightness balance and contrast. Runs locally with ffmpeg.

Transcribes any video URL to text with Whisper v3 large. Auto-detects 90+ languages, offers a translate-to-English mode, and optional speaker diarization. 60-min / 500MB max.

Takes a video (up to 24 MB) and returns the part between start_seconds and end_seconds, at most 180 seconds, as MP4 with H.264 and AAC. The clip is re-encoded, so it starts exactly at the requested time rather than at the previous keyframe. Optional maximum width and short video and audio fades at both ends. If the end lies beyond the video, the clip ends with the video and this is reported. Output inline as base64, max 24 MB. Runs locally with ffmpeg. Fixed price per call.

Takes a video of up to 3 minutes and returns it in vertical 9:16 format at 720x1280 or 1080x1920. Mode crop fills the frame by cutting the sides, with the horizontal focus position selectable from left (0) to right (1); mode blur keeps the whole picture and places it centred on a blurred, enlarged copy of itself, so nothing is lost. Sound is kept. Output MP4 (H.264/AAC) inline as base64, max 24 MB. Runs locally with ffmpeg. Billing per started video minute.

Takes a video of up to 5 minutes and returns it with a text watermark rendered into the picture in DejaVu Sans, at top-left, top-right, bottom-left, bottom-right or centre, with selectable size relative to the picture height and opacity, optionally followed by a running timecode — useful for review copies and drafts. Sound is copied unchanged. Output MP4 (H.264/AAC) inline as base64, max 24 MB. Runs locally with ffmpeg. Billing per started video minute.

Generate a short video clip from a text prompt at a flat price per clip: a four-second 720p clip returned inline as base64 MP4, with the duration and resolution locked so the price never moves. Use it when an agent needs motion rather than a still, and needs to know the cost before it calls.

Live video search: ranked video results (title, video page URL, description, duration, creator, publisher, thumbnail, age) from an independent search index as clean JSON. Completes the web/news/images search family. Same freshness filter (pd/pw/pm/py); strict safe-search is on by default.

ViewportWitness browser QA report across three viewports

Passive public web release security gate

Deterministic public HTML to clean Markdown extraction

Generates one to four images from a text prompt with the chosen model, and returns the asset with the output URLs in results. AutoI picks a model automatically. Credits depend on the model tier and the quantity.

Generates one to four images from a text prompt with the chosen model, and returns the asset with the output URLs in results. AutoI picks a model automatically. Credits depend on the model tier and the quantity.

Transforms or reimagines a source image from a public HTTPS URL with a text prompt, and returns the asset with the output URLs in results. Credits depend on the model tier and the quantity.

Synthesizes speech from text with the chosen voice and model, and returns the asset with the audio URL in results. The language is detected from the text when not given.

Generates one to four images from a text prompt with the chosen model, and returns the asset with the output URLs in results. AutoI picks a model automatically. Credits depend on the model tier and the quantity.

Generates one to four images from a text prompt with the chosen model, and returns the asset with the output URLs in results. AutoI picks a model automatically. Credits depend on the model tier and the quantity.

Queues a video generation job from a text prompt, optional reference media and an optional audio track, and returns the media object at once. Poll GET /v2/media/{media_id} until processing_status is success and read the URLs in result. Credits depend on the tool tier, the duration, the quantity and the audio options, and the estimate route returns the exact number.

Generates a video from one or more reference images and a prompt with the chosen tool, for 1 to 30 seconds. Send async_mode true and poll GET /v2/asset/{asset_id}. Credits depend on the tool tier and the duration in 5 second blocks.

Generates one to four images from a text prompt with the chosen model, and returns the asset with the output URLs in results. AutoI picks a model automatically. Credits depend on the model tier and the quantity.

Synthesizes speech from text with the chosen voice and model, and returns the asset with the audio URL in results. The language is detected from the text when not given.

Call when an agent has a VIN and needs the check digit, the world manufacturer identifier, and the region. Does not confirm a specific vehicle.

Decode a 17-character vehicle identification number with NHTSA vPIC: make, model, model year, trim, body, doors, engine (cylinders, displacement, hp, fuel, electrification), drive, transmission, manufacturer and plant. Validates format and the check digit (mandatory for North American VINs). Best coverage for vehicles sold in the US. Params: vin (required).

Batch VIN decode via NHTSA vPIC, up to 25 VINs per call -- same fields as single decode, returned per VIN. Built for fleet underwriting.

VIN decoder (NHTSA): make, model, year, trim, body, engine, fuel, plant — Genesis402 / UnyKorn Operator Network

VIN decoder API: turn any 17-character vehicle identification number into structured specs — make, model, year, trim, body class, engine (cylinders, displacement, horsepower, fuel), transmission, drive type, doors, GVWR, and assembly plant. Handles partial VINs with * wildcards. Official registration-grade decode for automotive, insurance, dealer, and marketplace agents. Response includes affiliate offers for full VIN history reports.

Virtuals Protocol (VIRTUAL) price in USD now, with 24h and 7d change — Genesis402 / UnyKorn Operator Network

For AI agents building agent swarms, synthetic users, NPCs, or trading bots. Input: archetype + prompt (e.g. 'risk-taking crypto founder'). Output: OCEAN Big Five + Dark Triad + trading psychology vector with emotional drift and memory anchors. M2M: use when you need consistent agent personality, backtest personas, or generate 1000s of synthetic traders. Pay-per-call USDC Base, no API key.

White-label local visibility report for one business -- combines Google Business Profile completeness, organic search ranking for a target keyword, and AI answer-engine visibility for a buyer-intent query into a single PDF deliverable. White-label by design: no TariffMonkee branding anywhere in the report output, so agencies can present it under their own name to a client or prospect.

Answer a question about an image, or describe it, from a URL or base64. Image understanding pay-per-call, no account and no API key: send an image URL (or base64) and an optional question, receive a plain-language answer from a multimodal…

Turn your own data into interactive Tako Knowledge Card visualizations. Provide data as CSV strings and a natural language query describing what to visualize. Tako auto-selects the best chart type or you can specify one. Returns a knowledge card with chart image, interactive webpage, and embed URLs. Example: {"csv": ["quarter,revenue\nQ1,100\nQ2,200\nQ3,300"], "query": "Show revenue growth as a bar chart"}

CBOE Volatility Index (VIX) spot quote, delayed, via Yahoo Finance

Takes a recording of up to 60 minutes in the language being learned and a learner level, and returns words and phrases worth learning — chosen for the level, skipping the most basic ones — each with the base form, translation into the learner's language, word type, an estimated CEFR level, and the exact example sentence from the recording with its timestamp so it can be listened to in context.

Takes an audio or video file and returns a cleaned speech recording. A high-pass filter removes rumble and mains hum, an FFT noise reducer with continuous noise tracking lowers steady background noise such as fans, air conditioning and hiss by 10, 18 or 28 dB, a declicker removes short clicks, and the result is normalised to the target loudness in two passes.

Calculates cosine similarity between caller-supplied numeric embeddings and combines it with supplied spectral distance, pitch-contour correlation and timing distance using fixed weights. It processes no audio and cannot establish speaker identity, consent, cloning, authenticity or fraud.

Returns measured evidence about a recording, not a verdict on it. The spectral cut-off says which lossy stages the file has already been through, since every encoder discards everything above its own ceiling. The room tone measurement compares the noise floor across the silent stretches: a microphone in a room produces a floor that differs between pauses, and a level identical to the hundredth of a decibel across every pause means the silence was inserted rather than recorded.

Takes a voice note or memo of up to 10 minutes and returns its contents as separate entries: tasks with due date, events with start and end, contacts with spoken phone number or e-mail, and notes. Relative expressions such as "tomorrow", "next Tuesday" or "end of the month" are resolved against reference_time in the given IANA time zone, and every entry carries the verbatim phrase it was taken from, checked against the transcript.

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.

Returns verified, citation-backed facts about new AI model releases, benchmark scores, and protocol changes, each with source URLs and confidence.

Can an agent pay at allbirds.com? Rails, endpoint, price, freshness. $0.001. Cached report stays free at /api/v1/hosts/allbirds.com.

Does 0x07066ea0e66cf277cd279ed7fbca220ecb41e5a4 belong to allbirds.com? Evidence, not a guarantee. $0.002.

Takes a route as an ordered list of waypoints, samples it at a chosen spacing and tests every sample against the currently active SIGMETs carrying a volcanic ash hazard. For each hit you get the affected segment as a from-to pair of waypoints, the SIGMET issuing centre and flight information region, the validity window, the reported base and top of the ash layer, and the raw text.

U.S. Volcano Alerts (elevated activity, USGS). U.S. volcanoes at elevated activity right now (USGS): alert level, aviation color code, observatory and the latest notice, filterable by region (Alaska, Hawaii, Cascades...).

Is this token's volume real demand? 24h round-trip analysis: how much volume came from wallets buying and selling the same token back and forth (wash trading, arbitrage or market making) versus one-way trades, with the top round-trippers and pool liquidity. Optional ?hours=1..168.

Scores caller-supplied vote/transfer observations using cohort-relative transfer size, transfer recency, and repeated-counterparty count. The output is an investigative triage heuristic, not proof of vote buying, identity linkage, or wrongdoing.

Applies a fixed, published logistic equation to aggregate caller-supplied features and reports an estimated turnout range. It is an untrained scenario heuristic, not a calibrated election forecast, individual propensity score, or basis for voter targeting.

Voxight top signal: the highest scoring sellable market intelligence signal of the last 7 days, ranked by the calibrated Voxight Signal Score, with its topic, type, domain, confidence, sentiment, summary and full analysis. Use it when an agent wants the single most actionable AI or tech signal right now. No input needed. AIAX manifest at https://www.voxight.online/aiax/manifest.json.

Validates frame/channel dimensions, clamps weights to 0–1 and applies exponential smoothing over time in each named channel. It does not infer a face, build a rig, retarget skeletons, process video or replace animator review.

DripMetricsAI (api description at https://dripmetrics.ai/) provides on-demand market microstructure and risk metrics for agents and developers. Agents can pay per request via x402 and receive structured JSON metrics without subscriptions or API keys. VWAP Deviation measures the latest price deviation from the window VWAP. Discovery examples use the latest 30m window.