Video.js v10 Beta Shrinks Players, Modularizes ABR With SPF
Mux announces the Video.js v10.0.0 beta, describing a ground-up rewrite that unifies work across Video.js, Plyr, Vidstack, and Media Chrome with a focus on smaller bundles and a more composable architecture for web video players. The post highlights an 88% reduction in the default bundle size versus Video.js v8 and introduces SPF (Streaming Processor Framework), a modular streaming engine approach that can significantly reduce ABR (e.g., HLS) footprint for simpler use cases. Mux notes that Mux Player users should not migrate yet, with feature parity and migration guidance targeted for mid-2026 and GA aimed for mid-2026, with ads support planned later in 2026.
Key Takeaways
- Bundle diet: Video.js v10’s default player is 88% smaller than v8’s default, with further savings via feature-based composition and presets (video/audio/background).
- ABR unbundled: v10 decouples adaptive streaming from the default build; it still works with HLS.js, Shaka, dash.js, and more when you need full power.
- New engine option: SPF (Streaming Processor Framework) composes “purpose-built” streaming engines—Mux claims ~38.5 kB minified / 12.1 kB gzip for a simple ABR engine, ~19% of v8+VHS for a basic HLS case.
- Developer ergonomics: first-class React/TypeScript/Tailwind support, swappable State/UI/Media components, and “ejectable” skins designed to be owned and customized (shadcn/ui-inspired).
- Timeline reality: beta APIs aren’t stable; feature parity and migration guides are planned for mid-2026, GA mid-2026, with ads support slated for later in 2026.
Why It Matters
Web players have quietly become 1MB “apps” that tax performance budgets—especially painful on mobile, emerging markets, and ad-heavy pages where every KB competes with measurement and monetization scripts. Video.js v10’s bet is that most teams don’t need the full ABR/DRM/ads kitchen sink for every embed, and that modular streaming (SPF) can make “good enough ABR” cheap to ship. If this sticks, the meme is “ABR by default, without the ABR tax,” pressuring other player/engine stacks to offer smaller, composable builds—while reminding enterprises that ads and migration timing (mid/late-2026) still dictate production readiness.
Read full article at mux.com