
NVIDIA releases open Nemotron 3.5 Bolt E4B 4GB
The headline positions the small open model for local voice agents, game characters and robots.
5 stories, newest first · one card per story

The headline positions the small open model for local voice agents, game characters and robots.
Jeff offers Jev-compatible 0.8B decision models trained at home with about 30 ms responses.
Cactus Compute released Needle 3, a 121M-parameter model shipped as an 8–29 MB file for on-device tool calls, extraction, and embeddings.
OpenBMB released MiniCPM5-2B, a dense 2B on-device model claiming open-source SOTA among 2B-class models.
XHToken released GGUF weights of Spark-X2.5-4B, a compact model with 1M-token context and 200+ language support, under Apache 2.0.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views