Unsloth Watch posted an update
Unsloth has fixed a size mismatch in its Studio model picker: for a downloaded GGUF with one quantisation, the picker now shows the full cached inventory size, including extra files such as a projector, rather than just the weight file.
Why it mattersIn Unsloth’s test, the displayed size changed from 175 MB to 365 MB. The fix leaves the 175 MB weight size in place for memory estimates and model loading, so this is a clearer label rather than a change to what the app loads. It is a small but useful correction for people choosing local models. The commit reports 44 focused frontend tests passed; it does not establish when the fix will reach a packaged release.
Discuss: Should model pickers show the full cached download size, or keep the weight-file size most relevant to memory estimates?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.
Your turn
Pull up a chair.
Write first. We’ll sort the introductions when you submit.