Synthesia connected its AI video platform to ChatGPT. You can now describe a video in the chat, get it generated with an avatar and voiceover, revise it by talking, search your existing Synthesia library, and trigger videos automatically from other connected tools — on every Synthesia plan.
2026-11-03 · by Moe Ameen
On November 3, 2026, Synthesia announced a ChatGPT plugin that brings its AI video platform directly into the chat, and said it is live on every Synthesia plan. Instead of switching to Synthesia's own editor, you describe the video you want inside ChatGPT, and the plugin generates it — avatar, voiceover, and supporting visuals — then lets you revise it in the same conversation. You can also search your existing Synthesia video library from the chat, and users without an account can create a free Synthesia account during setup.
ChatGPT gathers a few details before generating — it asks about visual style and length, then hands the brief to Synthesia to produce the clip. The integration exposes two levels of control. The default is a conversational mode built on Syren, Synthesia's prompt-to-video agent, aimed at the fast, designed output — explainers, how-tos, launch videos, and marketing — that can combine motion graphics, avatars, voiceover, music, and b-roll, all revised by chatting. If you specify Studio in the prompt, Synthesia switches to its precise lane: an avatar delivers a supplied script word for word, with brand kits and templates, and the result can be opened in Synthesia's editor for scene-by-scene changes. Studio is pitched at training, onboarding, and compliance content, where exact wording matters.
The second half of the launch is automation. Because the plugin sits alongside ChatGPT's other connected tools, a video can be triggered without anyone opening a chat at all — a workflow can run once, run on a schedule, or start when new material arrives in a connected app. Synthesia's examples include turning a Slack product-release post into a 30-second launch video and generating a welcome video for a new hire off a BambooHR check. To connect, you open ChatGPT's Plugins menu, search for Synthesia, and click Connect, then tag Synthesia in a new chat and ask for a video.
Synthesia's own pages also describe connecting through an MCP server that covers both ChatGPT and Claude, which is a separate route from the plugin. An older Synthesia "ChatGPT video generator" page describes an earlier, different workflow, so treat it as legacy. Synthesia did not publish specific plan limits or per-video pricing for the plugin beyond "every plan," so check the live plan details before you lean on it for volume.
The plugin ends the moment it is most useful to keep going: you now have a polished, on-brand Synthesia video sitting in a chat thread or your Synthesia library — and that is exactly where most creators get stuck, because a video in ChatGPT is not a post on a platform. [Kompozy](/) is the layer that takes over there. Drop the Synthesia MP4 into [Clipped Shorts](/glossary/clipped-short) and it cuts the explainer or launch video into vertical, platform-native cuts with word-synced burned-in captions, so the clip actually carries on muted feeds where Synthesia's generated voiceover plays silent. From the same brief you fed ChatGPT, Kompozy generates the formats the plugin never touches — a brand-exact [HyperFrames](/glossary/hyperframes) carousel, quote graphics, per-platform text, a blog post, and a newsletter — all held to one [Persona Brief](/glossary/persona-brief) voice.
Then [Autopilot](/glossary/autopilot) schedules and publishes the whole set across the eight social platforms plus blog and email, behind a per-post review gate. Synthesia's automation triggers one video into a library; Kompozy's autopilot turns one idea into a published week. And because Kompozy generates its own avatar video — [Persona Shorts](/glossary/persona-shorts) and HeyGen-based formats — you are not locked to a single vendor's render for the talking-head half. The clean division of labor: ChatGPT drafts, Synthesia renders the on-brand clip, and Kompozy clips it, multiplies it into every other format, and gets it live everywhere. To weigh the avatar tools themselves, see our [Synthesia](/ai-tools/synthesia) and [HeyGen](/ai-tools/heygen) pages.
It is an integration, announced November 3, 2026 and available on every Synthesia plan, that connects Synthesia's AI video platform to ChatGPT. You describe a video in the chat and Synthesia generates it with an avatar and voiceover, you revise it in the same conversation, and you can search your existing Synthesia video library — without leaving ChatGPT. It can also trigger videos automatically from other connected tools.
Open ChatGPT's Plugins menu, search for Synthesia, and click Connect; if you do not have a Synthesia account you can create a free one during setup. Then start a new chat, tag Synthesia, and ask for a video. Synthesia also documents an MCP-server route that works with both ChatGPT and Claude, which is separate from the plugin.
The default conversational mode is built on Synthesia's Syren agent and is aimed at fast, designed marketing video — explainers, how-tos, launch clips — combining motion graphics, avatars, voiceover, music, and b-roll, all revised by chatting. Studio, which you request in the prompt, is the precise lane: an avatar delivers your script word for word with brand kits and templates, editable scene by scene in Synthesia's editor, and is suited to training, onboarding, and compliance content.
No. The plugin generates and revises a Synthesia video and saves it to your library; it does not cut it into platform-native vertical shorts, caption it for muted feeds, or post it to any social account. To turn that clip into published posts across platforms, you still need a distribution and repurposing layer — for example, Kompozy clips the video, generates matching carousels, text, a blog, and a newsletter, then schedules and publishes them across nine platforms.