Nemotron Nano 9B v2
Compact hybrid reasoning model with controllable thinking budgets.
About
Nemotron Nano 9B v2 is a dense 9 billion parameter reasoning model from NVIDIA that blends Mamba2 and Transformer layers in a hybrid architecture. Released in August 2025, it supports a 128K token context window and lets developers control how much it thinks before answering, trading reasoning depth for latency.
It is aimed at chatbots, agents, and code assistance where a smaller footprint matters. The model is distributed under the NVIDIA Open Model License, which permits broad use with some conditions worth reading before commercial deployment.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 107; Trending score 72/100 from researcher assessment.