Z image turbo x402 REST usage
Z-Image Turbo: Eight-step turbo model. A cent per image.
Image, audio, video and speech generation or processing, including transcription and voice. Priced per call or per minute where a price is published at all.
Z-Image Turbo: Eight-step turbo model. A cent per image.
Call before an invoice, payout or onboarding uses a tax id from Brazil (CPF, CNPJ), Poland (NIP), France (SIREN, SIRET), Italy (partita IVA), Sweden (organisationsnummer) or India (PAN). Returns the id type and whether t...
Find copy-move (cloned) regions inside one image: block matching returns pairs of source and target boxes with the shift vector, matched block count and similarity. Copy-move detection by shift search and block verificat...
Locate double JPEG compression: JPEG-ghost sweep over recompression qualities returns the difference curve, earlier compression qualities of the whole image and regions that carry a different compression history. JPEG gh...
Decodes up to 2 minutes (mono, 16 kHz). Frame levels (20 ms window, 10 ms hop) give the overall 95th percentile and the noise floor (10th percentile); levels are reported relative to the 95th percentile of foreground fra...
Intrusive measure: both the degraded audio and the clean reference are decoded (mono, 10 kHz, up to 60 s). The degraded signal is aligned to the reference by cross-correlation (up to ±max_delay_s)....
Finds scene cuts with the ffmpeg scdet filter (scene_threshold on a 160 px wide copy), takes speech from caption cues when captions are supplied, otherwise detects it in the audio with Silero VAD v6.2 (or an energy detec...
Samples the video at sample_fps on a 320 px wide grey copy. For each candidate corner a box of area_share of the picture (default 1/6, the minimum signer size Ofcom concluded in its review of signing on television) with...
Probes all streams with ffprobe (video, audio with visual_impaired/descriptions disposition, subtitles with hearing_impaired/captions disposition and language), decides whether the media is audio-video, audio-only or vid...
Decodes the audio track (16 kHz mono), detects speech with Silero VAD v6.2 (MIT, pinned SHA-256) or an energy detector, and rasterises speech and caption display times on a 20 ms grid. The constant offset is the shift wi...
Decodes every frame at 64 px width and finds visual events as sudden rises of mean luma (a flash or white frame); finds audio events as onsets in a 1 ms energy envelope (rise above peak −20 dB afte...
Pass 1 runs ffmpeg 6.1 vidstabdetect (vid.stab 1.1, local motion fields), pass 2 vidstabtransform with the chosen smoothing window (frames), border handling (static optimal zoom, adaptive zoom, black or kept borders) or...
The audio window (up to 60 s, any audio file or the audio track of a video) drives an ffmpeg 6.1 showwaves waveform (mode cline, line, p2p or point, square-root amplitude scale, mono mix) in a band across the middle of t...
Each card is a full-frame ffmpeg 6.1 render: solid colour or linear two-colour gradient background, optional logo (PNG/JPEG/WebP, up to 22 % of the frame height, alpha kept), title in Liberation Sans Bold and subtitle in...
Samples the video at sample_fps on a 160 px wide grey image, splits it into runs without change (share of pixels changing by more than 24 grey levels at most change_threshold, against the previous sample and the run star...
Both videos (up to 300 s each) are sampled at sample_fps from their start, each sample is reduced to 64×64 grey and hashed with a 64-bit DCT perceptual hash (pHash method; aspect ratio ignored, so rescaling does not matt...
Both videos are decoded from start_s for duration_s (at most 60 s), converted to 8-bit YUV 4:2:0 and aligned frame by frame at the frame rate of the reference; the distorted video is scaled to the reference size with a b...
Every frame (at most 60 per second, up to 60 s and 2,400 frames) is reduced to a 96 px wide RGB image. Per pixel, relative luminance follows the WCAG 2.2 sRGB formula; a transition is an opposing change of at least 0.1 b...
Analyzes an agent-supplied image with AI and returns structured objects, text, normalized spatial metadata, insights and limitations.
Remove background from any image URL: returns a transparent PNG (base64) with the subject isolated with clean subject segmentation running locally on our hardware — no third-party image service. Use for product photos, p...
AgentDocs extracts supplier, invoice number, issue date, total, currency and text from one PDF invoice or receipt. JSON input with pdf_base64; maximum 1 MiB and 3 pages. Missing fields are null. Paid requests use a uniqu...
Image understanding and OCR with a vision LLM: describe an image, read text from screenshots, receipts and documents, or answer a question about it. POST a base64 PNG/JPEG (up to 4 MB) and an optional prompt.
Fast 1-step image conversion for AI agents: HEIC to JPG, JPG to PNG, PNG to WEBP, WEBP to JPG, SVG to PNG, AVIF to JPG, RAW to JPG, PSD to PNG, TIFF to PDF. Input formats include JPG, JPEG, PNG, WEBP, HEIC, HEIF, AVIF, G...
Fast 1-step audio conversion for AI agents: WAV to MP3, M4A to MP3, FLAC to MP3, OGG to MP3, OPUS to MP3, MP3 to WAV, AAC to MP3, WMA to MP3, AIFF to MP3. Input formats include MP3, WAV, FLAC, AAC, M4A, OGG, OPUS, WMA, A...