De courtes vidéos commentées qui montrent comment fonctionne ToRun : chat et recherche, images, vidéo et musique, workflows et assistants IA pour les lieux d'accueil.
A workflow is a ready-made recipe of AI steps. You pick a template, fill in a short form, and every step runs on its own until a finished video comes out. Here we follow one bedtime story video from the form to the result.
Let's open one workflow and look inside. A workflow is a graph: each box is a step, the arrows set the order, and steps that don't depend on each other run side by side. Your answers flow into the steps, some steps check their own work and run again, some wait for your approval, and every AI call is recorded with its own cost.
Every workflow, from a bedtime story to a research report, is built from the same few blocks. The AI step sends a prompt to a model and gets back text, an image, a voice, music or a video. Knowledge blocks bring in facts with their sources. Quality blocks check the work, and a human approval waits for you. Logic blocks shape the flow with branches and loops. Output blocks join everything into one video and save the result.
Found a workflow template that's almost right? Open it in the editor and make it your own. Click a step to change its prompt in the inspector, drag a new step in from the palette and connect it with arrows, then save it to your workspace. The template itself doesn't change; what you saved is your own workflow, and you run it from its form as before.
One question in, a cited research report out. We follow a question through the multi-source research template. An AI step splits it into smaller sub-questions. Searches of the web, academic sources and your own files run side by side, each only if you picked that source. Another AI step reads everything that came back and checks the sources against each other, then a citation step and a last AI step turn it into the report.
Write a story, pick a channel, and get back a finished episode. Inside the Story to episode template: a writer step turns your story into shots and lines of dialogue, every line is voiced in its character's own voice while music is made alongside, each shot is made with the characters speaking in sync with their lines, and a media step cuts it all together with captions, titles and your branding.
Before a workflow runs, its form shows the estimated cost for your exact choices. Every template card shows an estimated cost per run, and once people have used it, the typical cost of its recent runs. A preview is a low-cost draft that tells you exactly what it shortened. While it runs, every step shows its status, its time and its cost, every AI call is recorded with its own cost, and in the editor any step can have a budget cap.
Describe what you want, see the cost, and it gets made. Image generation with Auto picking the model, editing part of a photo, short videos priced by length and resolution, Enhance prompt for a clearer brief, songs with or without vocals, Text-to-Speech, and Moments templates that turn your photos into a video with a song. Every call is recorded with its own cost in your billing history.
In ToRun chat you just ask. Auto works out what each message needs and picks the best-value model that can do it; attach a photo and that turn moves to a model that can see it. Pick a model yourself and your choice is honoured. The assistant searches the web mid-answer and attaches its sources, uses tools like reminders and notes, runs Deep Research for bigger questions, and answers place questions with a map, nearest first. Your location is used only for that search.
A project groups related chats, files and memory, and every new chat in it follows the project's system prompt. Add PDFs, documents or text files under Knowledge and the assistant searches them in that project's chats. Memory works across all your chats: mention something worth keeping and you don't have to say it again. Everything it remembers is on your memory page, where you can edit it or make it forget for good.
Choose the language you speak and the one to translate into, and start talking: your voice is translated as you go. It works one direction at a time; tap swap and the other person answers in their own language, and the whole conversation can be saved as a note. Live Transcription writes down what you say in real time, Speech-to-Text turns a recording into text, and a video comes back with its transcript, its translation and subtitle files in both languages.
Share a chat with a link: choose when it expires and whether your uploaded files go with it, and anyone with the link sees a frozen copy without signing in. See how often it was viewed, and revoke it at any moment. Workflow results, canvases and media are shared the same way, with a preview card when the link is pasted. Your billing history lists every AI call on its own line, and the mobile app brings it all to iPhone, iPad, Android and Apple silicon Macs.
Accept the invitation, choose your own password and find your way around the venue panel: the Overview, the tabs, what your role lets you change, and the How it works button.
A visitor scans a QR code, asks a question in whatever language they write in, and gets an answer in seconds. There's no app and no sign-up. Answers come only from the venue's own material, directions start from where the code was scanned, and if the answer isn't there it says so instead of guessing. It never asks for personal data, and the venue's team sees what visitors asked and which answers were missing.
How a venue sets up its own AI assistant, from the documents it already has to the first answer a visitor gets. A setup checklist shows what's still missing. You add the venue's details and your brand, upload the documents you already have (PDFs, Word and Excel files, presentations, plain text, even a photo of a price list), mark the places on your floor plans and print a QR code for each point. Every answer comes only from your documents, and the demand report shows what visitors asked and what they couldn't find.
Upload your documents, write one right in the panel, edit it when something changes, give a campaign its end date, and remove what is out of date.
Let the assistant read your website every night, check changes before they are used, read again on demand, and add holidays and other special days.
Add a QR point, put it on the floor plan with its facing direction, print a designed poster in one or two languages, and show your codes on a lobby screen.
Ask your assistant what guests ask, see where each answer came from, set its tone and quick questions, and know how to stop it with Safe mode.
See what guests ask, where, when and in which language, find what they could not find, and close those gaps; or get the report by e-mail every Monday.
How a guest reports a fault without giving a name, how the report reaches your team at once, and how to acknowledge, resolve or dismiss it.
Invite colleagues by e-mail, pick the right role, send a new password link, and change or remove access when someone's job changes.
The assistant answers in the language of each question; set the visitor page's language, translate your names, and print posters in two languages.
Read the question allowance, what counts against it and its warning line, see where the questions come from, and find your plan and limits.
Everything ToRun does, in one video: creating images, video, music and voice; chat that picks the model, searches the web and does deep research; projects, your own files and memory; live translation and transcription; sharing and the billing history; ready-made AI workflows, from the form to the result, how to build and adapt them and what they cost; and AI assistants for venues, with their setup. Jump to any part from the chapters.
