b10416

ggml-org/llama.cppb10416Aug 13, 2026by github-actions[bot]same day as b10415

What changed since b10415

low risk

Web server now serves index.html with no-cache headers so the UI is no longer pinned to a stale build.

Improvements

  • index.html is now revalidated on each request via its existing ETag instead of being served as immutable, while unchanged builds still return 304.

Bug fixes

  • Fixed the web UI being stuck on an old build because the stable-named index.html inherited the long-lived immutable cache policy meant for hashed assets.

Upgrade notes

This is a single-commit server-side fix with no config or API changes; simply swap the binary. Clients that previously cached index.html will revalidate on their next request and pick up the current UI build. No migration steps are required.

Analyzed Oct 30, 2026

Upgrading?

Release notes summary

Fixes the web server to serve index.html with no-cache headers, ensuring users always receive the latest UI build.

Key Highlights

  • Serve index.html with no-cache (#27006)
  • Fixes pinned UI to old build issue

New Features

  • UI caching fix for the web server

Release Notes (excerpt)

<details open> server : serve index.html with no-cache (#27006) index.html was served with `max-age=31536000, immutable` like the hashed assets, but its name is stable while its contents change every build, so a cached copy pins the UI to an old build. It now revalidates via its existing ETag, which keeps the 304 for unchanged builds. </details> **Website:** - <https://llama.app> **macOS/iOS:** - macOS Apple Silicon (arm64) - macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED…

Published by ggml-org on GitHub.

Read the full release notes

Related Repositories