Parse LLM Markdown streams incrementally on the server or client
Incremental AST parsing beats full re-renders for long LLM markdown streams.
Zero-dep streaming markdown for the browser — Rust→WASM, 100% CommonMark 0.31 + GFM. Built to not melt the main thread on concurrent LLM streams.
Speculative closure parsing keeps LLM streams off the main thread while React melts.
Frontend developers building AI chat interfaces or real-time documentation tools
react-markdown · marked · commonmark.js
Incremental AST parsing beats full re-renders for long LLM markdown streams.
Event-stream parsing beats regex for streaming LLM tokens in real time.
12kb Markdown renderer with zero dependencies when react-markdown bundles 300kb+.
Token-level preview shows what LLMs actually parse before you ship prompts.
LLM-generated Markdown with embedded forms and approval gates instead of prose instructions.
Missing primitive: tolerant JSON parser for streaming LLM output, typed and <1KB.