Synthesia’s AI video agent — describe an on-brand video in a prompt and an agent writes it as code and renders it live in the browser, motion graphics, avatars, voiceover and all.
Last verified · 2026-10-26 · by Moe Ameen
Synthesia Syren is an AI video agent launched in beta in October 2026. Instead of building a video scene by scene on a timeline, you describe what you want in a chat prompt — or start from a preset or existing material you upload, such as a document, a deck, images, a reference video, or a YouTube link — and Syren plans and builds the whole thing: script, motion graphics, animation, avatars, voiceover, music, and b-roll. Its tagline is “On-brand video from one prompt,” and the output is a finished clip up to three minutes long, horizontal or vertical, produced in minutes.
What makes Syren technically distinct is how it renders. The agent writes the video as code, and an in-house rendering engine plays that code back live in the browser as the video is assembled — nothing to install, and you watch it take shape. You refine by chatting with the agent (“swap the avatar,” “change the script,” “restyle that scene”) rather than editing by hand. Coverage of the launch refers to the rendering engine as “Sirén”; Synthesia’s own pages describe an in-house engine without naming it, so treat the name as reported rather than official. Synthesia is positioning that engine as the foundation for a future of interactive AI video, though Syren today outputs a linear clip.
Syren is the fast, creative counterpart to Synthesia Studio, the company’s established editor for exact, repeatable training and compliance video. It uses Synthesia’s AI avatars and voices, both swappable in chat, while personal avatars and cloned voices still require Studio. Brand control is early in beta — Syren applies your colors now, with full brand kits (typography, animation styles) and learning a brand from your existing videos flagged as coming. It runs on credits, is free to try with launch credits (free output is watermarked), and is available to self-serve and Enterprise customers. Treat specs and pricing as a beta snapshot and verify on synthesia.io.
Think of Syren as the designer and [Kompozy](/) as the rest of the studio plus the distribution desk. Syren is excellent at one job: turning a brief into one polished, on-brand video. But one three-minute clip is a single asset, and a content week is dozens of them across formats and platforms. That is the multiplier Kompozy adds — one Syren render becomes the seed for everything else. Export the Syren MP4 and feed it to Kompozy’s [Clipped Shorts](/glossary/clipped-short), which cuts the long clip into vertical, platform-native shorts and burns in word-synced captions — the step that makes a Syren video carry on the feeds where its generated voiceover plays muted.
From the same idea, Kompozy generates the formats Syren does not touch: a brand-exact [HyperFrames](/glossary/hyperframes) carousel, quote graphics of the strongest line, per-platform text posts, a blog, and a newsletter — all written to one [Persona Brief](/glossary/persona-brief) so the copy matches the video’s voice. Kompozy’s own [Persona Shorts](/glossary/persona-shorts) and [avatar video](/glossary/avatar-video) can add a talking-head segment, so you are not tied to one vendor’s render. Then [Autopilot](/glossary/autopilot) schedules and publishes the whole set across the eight social platforms plus blog and email, behind a per-post review. Syren makes the on-brand video; Kompozy turns that one render into a published, multi-format week.
Syren is Synthesia’s AI video agent, launched in beta in October 2026. You describe a video in a prompt (or start from a preset or uploaded context), and an agent writes it as code and renders it live in the browser — building script, motion graphics, animation, avatars, voiceover, music, and b-roll into an on-brand clip up to three minutes long. You refine it by chatting with the agent rather than editing a timeline.
Studio is Synthesia’s editor for exact, repeatable content — training, compliance, onboarding — where you control every script line and scene. Syren is the prompt-driven agent for fast, creative, designed video such as launches, explainers, and ads. Personal avatars and cloned voices still require Studio; Syren uses Synthesia’s stock avatars and voices, swappable in chat.
Not today. Syren outputs a linear clip that plays as a fixed file. Synthesia is positioning the code-native rendering engine (reported as “Sirén”) as the foundation for interactive AI video, but that is a stated direction rather than a current Syren feature. Interactive products like Interactive Avatars and Roleplay Sessions are separate parts of Synthesia.
Syren is in beta and credit-based. It is free to try with launch credits for a first video (free output carries a watermark), and it is available to Enterprise customers whose admins can disable it. Synthesia says a two-minute video with a couple of rounds of edits typically uses roughly 1,000 to 2,000 credits. Verify current pricing on synthesia.io.
No. Syren renders a video file and stops — captioning, reframing, and posting are manual. A content engine like Kompozy covers that half: it clips the Syren video into captioned vertical shorts, generates carousels, text, a blog, and a newsletter from the same idea, and schedules the set across eight social platforms plus blog and email.