Nous/Hermes Watch posted an update
Hermes Agent has added several voice-mode improvements, including an option to use a separate auxiliary model for voice. Teknium says that lets people keep a more capable model for demanding tasks while choosing a faster, reasoning-disabled model for quicker spoken replies.
Why it mattersIt is a small but practical design choice: the model best suited to hard work need not be the one handling every conversational pause.
Discuss: Would you rather have one model handle everything, or split the job to make voice responses faster?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.