INFRO Studio

INFRO Studio (studio.infro.io) is a workspace for making images, video and audio as a team. It runs on the same organization, members, roles, provider connections, projects, budgets and plan as the console. Every generation it makes is a gateway request on one of your organization's own provider connections, so your provider bills you for it directly, under your own contract. INFRO charges its subscription only, with 0% markup on model spend.

What Studio does

• Create. Studio opens on Create: choose image, video or audio, then the task (generate, edit or upscale an image; generate, animate an image, extend, restyle, upscale or lip-sync a video; voice, sound effects or music), then the model, then the provider that runs it — Auto, or one of your connections serving that model, each with its rate. Inputs are dropped in, uploaded or picked from the Library, the side column shows only the settings that model takes on that provider, and the estimate shows before Generate.
• Storyboard. Scenes and shots, each shot with its versions, takes and the pick. A shot moves through four stages — keyframe, animate, sound, finish — and Run all plans every step of a sequence with an estimate before anything runs.
• Cast. Characters, locations, styles, props and voices reused across shots and models, with the references each model takes chosen for it. A character built from photos of a real person needs its creator's attestation that they have that person's permission; it is recorded, and a run without it is refused.
• Director. Turns a written brief into a cast, shots and prompts, on a text model served by one of your own connections.
• Recipes and Flow. Multi-step recipes (a talking character, upscale and finish, a music video cut to the beat) and Flow, a canvas of the steps. A run that stops can be resumed, reusing the results it already made.
• Voice, sound and finishing. Speech with a voice picker and preview, sound effects, music, lip-sync, image and video upscale, and captions.
• Export. A package with the chosen takes, keyframes, sound and captions, timelines for Final Cut Pro (FCPXML), DaVinci Resolve (EDL) and Premiere Pro (XML), and a manifest.json that records the model, provider, prompt, settings, estimated provider cost and lineage of every file.
• Review. Comments at a moment in a video or on a region of an image, approve or request changes on a shot, and view-only share links that expire within 30 days and can be revoked. A share link shows titles and pictures, never a prompt, model, provider or cost.
• Edit from a comment. On an image, the comment box can also make what is written: Edit sends the picture and the words to an image model on one of your connections, as a new picture beside the original. With a region marked, a model that takes a mask (GPT Image) changes only that region; any other model is told where the region is in words, and the comment box says which before you run it, with the model, the provider and the estimated provider cost. A comment already left can be turned into the same edit.

How it relates to the API

/studio/v1 is the API the Studio app itself calls. It takes the signed-in member's session, answers only Studio's own origin, and applies the member's role: owners, admins and developers make, upload and export; analysts and finance view, comment and review. It is not a gateway-key API and has no SDK. To automate the same work from an application, call /v1 with a gateway key — Studio's runs go through the same /v1 implementations, so the model checks, routing, failover, budgets, model allowlists and guardrails you configure apply to both.

Cost before every run

Before a run starts, Studio shows its plan and the estimated provider cost of every step, and what is left of the member's own monthly limit and the organization's budget. The estimate is what your provider is expected to bill you, from your negotiated rate where you have entered one and the vendor's list rate otherwise; where no rate is known the step is shown as unpriced, never as $0. Budgets and member limits are enforced on Studio's requests exactly as on any other request. Your provider's invoice is authoritative.

The same estimate is available to an application as POST /v1/estimates.

Storage and retention

Studio stores nothing until an owner or admin of the organization turns its storage on; who did and when is recorded in the audit log. Once it is on, Studio keeps uploads, generated media and, for each result, its prompt, model, provider connection, settings and estimated provider cost. Studio storage is separate from request payload logging: turning one on never turns on the other. Generated media is stored as the provider returned it, so provenance marks the provider adds stay in the file.

• An organization can choose a shorter window than its plan's. Every item shows the date it is kept until, and the people who made items leaving within a week get one email listing them.
• Deleting moves an item to Trash for 7 days, then it is purged. Downloads and exports are always available while an item is kept.
• Turning storage off stops new storage and either deletes everything now or lets items expire on schedule — your choice.

Reports and takedown

Any result in Studio, and anything shown through a share link, can be reported. A person at INFRO reviews each report, and content that breaks the acceptable use policy is taken down within 48 hours: it stops being served everywhere, including share links.

The /v1 capabilities Studio added

Studio needed media capabilities the gateway did not have. They are on /v1 for every application, each checked against what the model takes on each of your providers before any call:

• Reference images, last frames, negative prompts, extend and video-to-video: images and video.
• Upscale, lip-sync, sound effects and music: image upscale, video upscale and lip-sync, sound effects and music.
• Cost estimates before a request: billing.
• Long renders run as async jobs with signed webhooks. Studio's own runs do not send webhooks.