Stream

Stream provides developer-friendly APIs and SDKs for real-time chat, video, audio, feeds, and AI-powered moderation—powering in‑app communication for 1B+ end users.

Vision Agents is Stream’s open-source framework that helps developers quickly build low-latency vision AI applications. Since its initial launch, the project has expanded with additional plugins, better model support, and major improvements to latency, audio, and video handling. It ships with out‑of‑the‑box integrations and broad model support (e.g., OpenAI Realtime, Gemini), with ongoing improvements to latency, audio, and video handling. Deepgram plugs in natively as the STT provider to deliver fast, accurate real‑time transcription (and diarization) inside Vision Agents workflows.

With v0.2, Vision Agents continues to evolve, bringing new plugins across avatars, VLMs, TTS, and more, making it even easier to build powerful multimodal AI features with minimal code.