BenchMark'd
No ratings yet
Be the first to review this model
NVIDIA's compact agentic model with hybrid latent mixture-of-experts architecture, delivering 4x higher throughput than Nemotron 2 Nano. Designed for efficient multi-agent system deployment at the edge. Ideal for orchestrating reliable agent workflows with minimal compute requirements.
Released
December 15, 2025
Parameters
30B (MoE, 3B active)
Context
128K
Pricing
Open Source