Synthesia's Syren builds a finished video from a prompt
The tool renders in the browser at up to three minutes and is credit-based, with launch credits covering every customer's first video.

Key takeaways
- Synthesia launched Syren in beta on October 9, 2026, a tool that builds a finished video from a prompt, a preset or a supplied branded document, image, video, PowerPoint or YouTube link.
- Syren renders in the browser at up to three minutes, horizontal or vertical, and is credit-based; launch credits let every Synthesia customer make a first video.
- Syren runs on a video rendering engine Synthesia built in house that has powered motion graphics inside Synthesia since June.
- Syren joins Roleplays, Surveys and the Synthesia editor on the platform, and will soon learn a company's brand from its existing videos.
Synthesia launched Syren in beta on October 9, a tool that builds a finished video from a prompt, a preset or a document, image or video you already have.
The London-based company announced the launch in a blog post by chief technology officer Peter Hill. Syren writes the video as code, then a rendering engine plays it back live, so the video takes shape as it is built, Hill wrote.
A creator can start from a sentence, drop in a branded document, image, video, PowerPoint or YouTube link, and Syren assembles motion graphics, animation, avatars, voiceover, music and b-roll. A new editor inside Syren adds playback, a transcript view and chat-based editing, so changes come from asking rather than starting over.
Synthesia is the London avatar-video platform founded in 2017 and used by over 90% of the Fortune 100, according to Wikipedia. Syren runs on a video rendering engine Synthesia built in house. That engine has been powering motion graphics inside Synthesia since June, and Synthesia's update notes name Motion Graphics as the first capability the model powers.
Videos render in the browser, with nothing to install, and run up to three minutes, horizontal or vertical, the post says. Syren is credit-based, and launch credits let every customer make a first video. Enterprise administrators can switch it off at the organization level.
Syren joins Roleplays, Surveys and the Synthesia editor on the platform. It is not Synthesia's first agentic product: the company launched Sessions, a platform for two-way conversations with avatars, on October 1. The editor stays the place for content that has to be exact and repeatable, including compliance training and onboarding.
Hill described Syren as "a creation experience where the AI agent doesn't assist the editor, it is the editor." Chief executive Victor Riparbelli wrote on LinkedIn that Syren is "now live and FREE to try," and that it runs in the browser or through a Claude connector.
Syren will soon learn a company's brand from its existing videos, matching graphics, motion style, pacing and the rhythm of its edits, the post says.
Syren is available in beta to self-serve, free and all Enterprise customers, through a Syren button in the left-hand navigation of Synthesia.
Sources
- synthesia.io - launch post by CTO Peter Hill: beta date, inputs, outputs, engine powering motion graphics since June, three-minute in-browser render, credits, availability
- synthesia.io - Syren product page: example outputs and a 16:9 1920x1080 prompt
- synthesia.io - confirms SyRen as Synthesia's in-house rendering model and the first capability it powers
- linkedin.com - chief executive Victor Riparbelli's launch post: live and free to try, browser and Claude connector
- unite.ai - Sessions launched October 1, 2026, context that Syren is not Synthesia's first agentic product
- en.wikipedia.org - London headquarters, founded 2017, used by over 90% of the Fortune 100