SeedRealtime
ByteDance's Seed team has released a voice and video model that can listen, watch and talk at once, reached through hosted access rather than a download.
Key facts
- Released
- Aug 5, 2026
- What it does
- Text and chat: Writes, answers questions and holds a conversation.Sees images: Understands pictures, screenshots and charts you share with it.Watches video: Understands video clips you share with it.
View moreView less
- How
- Only in the maker's own app or website
- Announced as
- “SeedRealtime Audio-Visual Full-Duplex LLM Released: Toward Omni-Modal Natural Interaction”
From trckr's archive of releases before it started watching. Date and details from Epoch AI's database of AI models (CC BY 4.0).
Compare SeedRealtime with
It came out 43 days after Doubao Seed Evolving, the lab's previous model.
The details
SeedRealtime is an audio-visual full-duplex Language model (LLM): A model trained on huge amounts of text to read and write. The engine behind chatbots like ChatGPT, Claude and Gemini. from ByteDance Seed that handles speech, image, and video understanding in real time, available only via hosted access with no public API: How developers plug a model into their own apps and pay per use. A model that's "in the API" can be used by programmers, even if it isn't in a chat app yet..
Get alerts like this the moment they happen.
From ByteDance Seed and every other major lab.