• VS Code extension 0.6.4
    ci / build (push) Canceled after 0s
    Pre-Release

    Harvmaster released this 2026-10-04 12:11:42 +11:00 | 0 commits to main since this release

    • Accepting a suggestion asks for the next edit immediately (setting nextEdit.afterAccept, on by default). After
      Tab on a completion or a next-edit, the follow-up is requested right away instead of after the typing pause, so
      accepted next-edits chain (rename → next use → …).

    Install: download local-autocomplete-0.6.4.vsix below, then code --install-extension local-autocomplete-0.6.4.vsix (or Extensions → … → Install from VSIX). See the README.

    Downloads
  • VS Code extension 0.6.3
    ci / build (push) Canceled after 0s
    Pre-Release

    Harvmaster released this 2026-10-04 12:06:34 +11:00 | 1 commits to main since this release

    • Fix: model choice didn't stick in managed mode. The daemon's saved state (last model used) overrode both the
      model picked in the menu and a new default from a release, so it kept restarting on the old model (e.g.
      stage-c). A changed default now wins; switching through the daemon API still persists as before.

    Install: download local-autocomplete-0.6.3.vsix below, then code --install-extension local-autocomplete-0.6.3.vsix (or Extensions → … → Install from VSIX). See the README.

    Downloads
  • VS Code extension 0.6.2
    ci / build (push) Canceled after 0s
    Pre-Release

    Harvmaster released this 2026-10-04 11:54:52 +11:00 | 2 commits to main since this release

    • Next-edit is ~3–5× faster: the next-edit server uses n-gram speculative decoding (the rewritten region is mostly
      a copy of the current one, so drafted tokens are verified in batches). Same output, since decoding is greedy.
      Rewriting a 13-line region on a laptop CPU: ~2.1 s → ~0.7 s (~0.4 s when the prompt is cached).
    • Next-edit timings in "Show Last Prompt" now split TS errors / prompt / generation time.

    Install: download local-autocomplete-0.6.2.vsix below, then code --install-extension local-autocomplete-0.6.2.vsix (or Extensions → … → Install from VSIX). See the README.

    Downloads
  • VS Code extension 0.6.1
    ci / build (push) Canceled after 0s
    Pre-Release

    Harvmaster released this 2026-10-04 11:23:31 +11:00 | 3 commits to main since this release

    • New default model stage-d, with stage-e for next-edit. Both download automatically (~530 MB each); the
      daemon runs stage-e in a second llama-server, only for next-edit. stage-d closes blocks (}) and predicts the
      next line far better (held-out exact match: block close 0.27 → 0.79, next line 0.01 → 0.18), and sees TypeScript
      errors near the cursor. Results in models/README.md.
    • Fix: a "no edit" answer from a next-edit model could be applied as deleting the suggested region. llama-server
      drops the model's no-edit token, so it arrived as empty text. Next-edit models are only enabled from this version on.
    • Manifest: nextEditModel (companion model) and nextEditFormat (replaces nextEdit, which older versions ignore
      from now on).

    Install: download local-autocomplete-0.6.1.vsix below, then code --install-extension local-autocomplete-0.6.1.vsix (or Extensions → … → Install from VSIX). See the README.

    Downloads
  • VS Code extension 0.6.0
    ci / build (push) Canceled after 0s
    Pre-Release

    Harvmaster released this 2026-10-04 04:21:41 +11:00 | 7 commits to main since this release

    • Next-edit suggestions (with models trained for it, e.g. the upcoming stage-e): after you pause typing, the model may suggest an edit around the cursor, such as removing a leftover ) or updating a renamed variable's uses. Single-line changes show deleted text struck through and inserted text as ghost text; multi-line changes are highlighted with the diff on hover. Tab applies, Esc dismisses. Setting: nextEdit.enabled.
    • Token healing: completions that start mid-word (this.connections.del▮) now finish the word instead of skipping it.
    • Suggest on cursor move (triggerOnCursorMove, off by default): with next-edit models it predicts what you came to change; otherwise it only fires on empty lines, line ends and lines with errors.
    • Pre-warm on focus (prewarmOnFocus, off by default): fills the model cache when you switch files, so the first keystroke is fast.
    • Show Last Prompt command: the exact prompt the model saw, its raw output and timings.
    • Update check against the Releases page (daily; checkForUpdates), with one-click install.
    • Edit recorder also records VS Code's diagnostics (TS, Vue, ESLint) for recorded files.
    • Daemon: model profiles carry the prompt format they were trained with (contextFormat, errors, nextEdit), so newer models get TS errors and namespace-import symbols in their context.

    Install: download local-autocomplete-0.6.0.vsix below, then code --install-extension local-autocomplete-0.6.0.vsix (or Extensions → … → Install from VSIX). See the README.

    Downloads
  • Harvmaster released this 2026-10-04 01:42:25 +10:00 | 9 commits to main since this release

    • New repository layout (models/, extensions/vscode/, daemon/, shared/). The model manifest moved to models/manifest.json, and the default localAutocomplete.modelManifestUrl follows it.
    • Publisher is now harvmaster (extension id harvmaster.local-autocomplete). Uninstall the old local.local-autocomplete first. The model is downloaded once more, into the new extension's storage.
    • MIT license, repository links, versioned VSIX file names.

    Install: download local-autocomplete-0.5.0.vsix below, then code --install-extension local-autocomplete-0.5.0.vsix (or Extensions → … → Install from VSIX). See the README.

    Downloads