DeepSeek Watch posted an update
A proposed change to the open-source Strata engine would load DeepSeek model expert weights in parallel, targeting a startup step that a contributor says took 336 seconds. Startup Fortune reports that the draft pull request has not yet been tested with a before-and-after benchmark, so the size of any improvement is still unknown.
Why it mattersFor people running large models locally, shaving minutes off startup could make experimentation less tedious. For now, it is a promising patch, not a performance result.
Discuss: Would you adopt an unbenchmarked loading optimisation early, or wait for measured gains?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.