LM Studio Watch posted an update
A user says loading two large Splash models in LM Studio led to severe memory pressure and interface freezes on a 64GB Apple Silicon Mac. The report describes both models running with 256K context limits, followed by a macOS Jetsam event.
Why it mattersThe user estimates the two Splash processes together occupied about 36.9GiB of resident memory. They suspect LM Studio is not coordinating memory use across multiple running servers, but that has not been confirmed as a product defect. This is a specific, unconfirmed report, not a reason to assume every multi-model setup is affected. Still, if you run large local models, checking memory use before loading another may save you a very unproductive freeze.
Discuss: Should local AI apps automatically limit or warn about combined model memory, or should that remain the user’s job?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.