What changed
Nous Research released DeepHermes-3-Llama-3-8B-Preview with weights and a system-prompt switch between ordinary responses and long-form reasoning. The model also targeted function calling, JSON output, and local serving.
Why it matters
It made a hybrid reasoning interface available in a small downloadable model, useful for studying routing, controllability, and local deployment.
What to do
Run it locally as an experimental worker for bounded reasoning or structured-output tasks, and compare reasoning-on versus reasoning-off on the same tests.
Caveat
Nous explicitly called it a preview with quirks. Benchmarks were self-reported, the base licence was the Llama 3 community licence, and visible chain-of-thought can be misleading or sensitive.