Pick your stack

Two questions decide what to install.

1. Which framework?

You're writing…	Use this adapter	Page
Plain TypeScript / Vite / no framework	@forinda/video-sdk-core directly	core
React 18+	@forinda/video-sdk-react (hooks + <VideoView>)	react
Vue 3.4+	@forinda/video-sdk-vue (composables + <VideoView>)	vue
Web Components / Lit / Stencil / no framework	@forinda/video-sdk-elements (4 custom elements)	elements

The framework adapters are thin wrappers around @forinda/video-sdk-core — same lifecycle, same events, just expressed in your framework's idiom.

2. Mesh or SFU?

Scale / shape	Use
1-on-1 / small group (≤ 6–8 peers, full mesh)	Mesh — definePublisher + defineViewer. Default.
1-publisher webinar (1 → N viewers, N > ~8)	SFU — defineSfuPublisher + defineSfuViewer via the LiveKit adapter.
Town-hall (2-3 publishers, many viewers)	SFU. Same reason.
Recording at scale, server-side composited	LiveKit egress directly (out of scope for the SDK).

The SFU adapter has the same Publisher / Viewer shape as mesh, so swapping is one factory call. See SFU integration for the full mesh→SFU migration.

Then add what you need

• Chat / presence / raise-hand — defineRoomChannel (or room.channel() if you're using defineRoom). Rides the same WebSocket as media. See Patterns.
• Recording — defineRecorder(stream, { timesliceMs: 1000 }). Ships chunks to your uploader.
• Screen share — getDisplayMedia() then feed it to a second Publisher with a different room name (or replace the camera track on the existing one).