Nemotron 3 Super
Provider：NVIDIA
NVIDIA Nemotron 3 Super is a 120B-parameter open hybrid MoE model, activating just 12B parameters for maximum compute efficiency and accuracy in complex multi-agent applications. Built on a hybrid Mamba-Transformer...
Key specs
- Context：1M
- Max output：16.4K
- Tokenizer：Other
- Released：2026-03-11
Token pricing
- Input price：$0.085 / 1M tokens
- Output price：$0.400 / 1M tokens
- Blended price：$0.164 / 1M tokens
Modalities
Text
Use Cases
- Deep Reasoning & Analysis：Process long legal docs, academic papers, financial reports; multi-hop reasoning and evidence-chain tracing across huge contexts.（≈ $0.0032/call）
When to pick this model
Signals derived from public pricing and spec fields — not a benchmark. Validate with official docs and your own eval.
- Low input + output price: solid fit for high-frequency calls and batch processing.
- Low blended price: budget-friendly for fleet-scale deployment.
- Long context (>=1M tokens): well-suited for long-doc QA, code-base analysis, multi-turn agents.
FAQ
- What is the token pricing of Nemotron 3 Super? Input price $0.085 / 1M tokens, output price $0.400 / 1M tokens, blended about $0.164 / 1M tokens. Refer to NVIDIA’s official page for the exact rate.
- What context window does Nemotron 3 Super support? Context window is 1M, max output about 16.4K, suitable for long documents, multi-turn chat and complex reasoning.
- What scenarios is Nemotron 3 Super best for? Based on its capability and pricing, Nemotron 3 Super fits best: Deep Reasoning & Analysis. See the Use Cases section for details and cost estimates.
- What input/output modalities does Nemotron 3 Super support? Supports Text modality.
- Is Nemotron 3 Super a free model? Nemotron 3 Super is billed per token, not a free model.
- Which provider offers Nemotron 3 Super? Nemotron 3 Super is offered by NVIDIA.
More models from NVIDIA
Full info：Nemotron 3 Super