Nemotron 3 Nano
Released
2026-03-16
Last refreshed
2026-10-26
Status
Researched 6d ago
Open weightsCommercial use: permittedRAGAgentsLong contextJSON / Tool use
Nemotron 3 Nano is a released rag, agents, and long context model with open-weight and 256K context; evaluate it while provider pricing coverage matures.
NVIDIA AI releases · 27 in the last 12 months · this family litChangelog →
Specifications
- Family
- Nemotron 3
- Released
- 2026-03-16
- Context
- 256K
- Parameters
- 3.97B
- Architecture
- Mixture of Experts
- Specialization
- general
- Openness
- Open weights
- License
- NVIDIA Open ModelPermitted
- Weights
- Available
- Code
- Unknown
- Training
- Pretrained
Created by
Pricing
No tracked provider pricing is available yet.
About
NVIDIA's lightweight 3.97B parameter model optimized for edge deployment with FP8 quantization (W8A8 mixed precision). Designed for agentic AI applications including gaming NPCs, local voice assistants, and IoT automation. Supports instruction following, tool use, and hallucination avoidance. Strong performance on BFCL, IFBench, IFEval, HaluEval, RULER, Tau2, AIME25, MATH500, GPQA-D, and LiveCodeBench.
Providers
No active provider routes are tracked for this model yet.
Capabilities
JSON / Tool use
Benchmark peer barsfor RAG
RULERRank 5 of 31
Benchmark scores(1)
Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
Migration checks
No linked migration route is available for this model yet.