Use cases

Markstream is designed for scenarios where Markdown arrives incrementally. Here are the most common use cases.

AI chat and LLM streaming

• Vue AI chat Markdown renderer — build Vue 3 and Nuxt chat surfaces for LLM token streams
• LLM token stream Markdown — batch and render chunks from SSE, WebSocket, fetch streams, or custom transports
• AI chat streaming Markdown — render LLM token streams in Vue, React, Svelte, and Angular
• SSE and WebSocket streaming — handle Server-Sent Events and WebSocket Markdown output
• Incomplete Markdown renderer — keep unclosed fences, partial tables, math, and HTML stable while a response is still streaming
• Streaming code blocks — render code fences that arrive token by token with an enhanced stream-diffs surface or plain pre fallback

Progressive heavy blocks

• Streaming Mermaid and KaTeX — render diagrams and math incrementally during streaming

Performance at scale

• Long AI responses — virtualized rendering for 50 KB+ AI output

Quick reference by framework

Framework	Package	AI chat	SSE/WS	Mermaid/KaTeX	Long docs
Vue 3 / Nuxt	markstream-vue	✅	✅	✅	✅
React / Next.js	markstream-react	✅	✅	✅	✅
Svelte 5	markstream-svelte	✅	✅	✅	✅
Angular	markstream-angular	✅	✅	✅	✅
Vue 2	markstream-vue2	✅	✅	✅	Partial

See also

• Compare Markstream with alternatives
• Framework-specific guides
• Installation by scenario