John O'Reilly on Mythryl: “Tested the new NVIDIA Bolt E4B on-device model on a few of my repos…”

Tested the new NVIDIA Bolt E4B on-device model on a few of my repos and performance was basically identical to standard Gemma 4 E4B. It’s tuned for agentic work (tool use/multi-turn agent tasks), but I didn't see much difference. research.nvidia.com/labs/nemotro...

research.nvidia.comNVIDIA Nemotron 3.5 Bolt E4B 4GBWe are releasing NVIDIA Nemotron 3.5 Bolt E4B 4GB, an open small language model optimized for speed and intelligence per GB in conversational agents.

25/10/26, 5:52 pm0 reposts4 likes

Post