vsmdsvex.

Build-Time Preprocessor vs Runtime Component

get started ↗ examples all comparisons

FIG-001

SHEET 01 / 07

FIG-002 / OVERVIEW

at a glance.

Compare MDsveX and @humanspeak/svelte-markdown: build-time .svx preprocessing versus runtime Svelte 5 markdown rendering, caching, and HTML control.

FIG-003 / FEATURE MATRIX

side-by-side.

Every surface that matters, compared without spin.

feature	@humanspeak/svelte-markdown	MDsveX
Svelte 5 Native	yes	yes
TypeScript Support	yes	Partial
Runtime Rendering MDsveX is build-time only	yes	no
Dynamic Content Cannot render user-supplied markdown	yes	no
Custom Renderers	yes	yes
Token Caching Not needed — parsed at build time	yes	no
LLM Streaming Mode MDsveX is build-time only, so there is no runtime markdown streaming mode.	writeChunk() / resetStream()	no
Streaming HTML Output MDsveX compiles at build time — there is no runtime stream that could resolve partial HTML blocks across chunks.	Partial blocks reconcile when </tag> arrives	no
HTML Tag Control	84 tags with allow/deny	Via rehype plugins
Markdown in Components	yes	yes
Components in Markdown	no	yes
Frontmatter MDsveX has built-in frontmatter parsing	no	yes
Math (KaTeX)	Built-in extension (markedKatex)	Via remark-math + rehype-katex
Diagrams (Mermaid)	Built-in extension (markedMermaid)	Via rehype-mermaid
GitHub Alerts	Built-in extension (markedAlert)	Via remark-github-alerts
Code Highlighting	Via marked extensions	Built-in (Shiki/Prism)
Remark/Rehype Plugins	no	yes

FIG-004 / STRENGTHS

where each shines.

▣ svelte markdown 13

+Svelte 5 runes-native — built for runes, not retrofitted
+TypeScript-first with full type safety
+Built-in token caching (50-200x faster re-renders on repeated content)
+LLM streaming with imperative writeChunk() / resetStream() API
+23 markdown renderers + 84 HTML tag renderers — every override is a Svelte snippet
+First-class extensions: KaTeX math, Mermaid diagrams, GitHub alerts, footnotes
+Built-in XSS protection — protocol allowlist, event-handler stripping, attribute sanitization
+Allow/deny utilities (allowHtmlOnly, excludeRenderersOnly, etc.) for fine-grained control
+Drop-in component — works anywhere in your Svelte app
+Renders dynamic/user-supplied markdown at runtime
+No build step required — works with any markdown string
+Math, Mermaid, alerts, and footnotes ship as first-class extensions — no remark/rehype pipeline required
+Simpler mental model — just pass a string, get rendered output

▢ mdsvex 6

+Largest Svelte markdown ecosystem (~3,000 GitHub stars)
+Build-time optimization — zero runtime parsing cost
+Use Svelte components directly inside markdown files
+Rich plugin ecosystem via unified/remark/rehype
+Built-in code syntax highlighting
+Frontmatter parsing out of the box

FIG-005 / LIMITATIONS

where each falls short.

▣ svelte markdown 5

−Smaller community (newer project)
−Runtime-only (no build-time optimization)
−Cannot embed Svelte components inside markdown content
−No frontmatter support (parse separately if needed)
−No built-in code syntax highlighting (use a marked extension)

▢ mdsvex 4

−Cannot render dynamic/user-supplied markdown
−Content must exist at build time as .svx files
−Adds build complexity (preprocessor configuration)
−Not suitable for CMS content, user input, or API-fetched markdown

FIG-006 / VERDICT

the honest call.

Choose MDsveX for static content sites, blogs, and documentation where markdown is known at build time. Choose @humanspeak/svelte-markdown when you need to render dynamic markdown — CMS content, user input, API responses, or any scenario where the markdown string is not known until runtime.

FIG-007 / MORE