Model Library
NVIDIAReleased March 19, 2026Nemotron Cascade 2 30B A3B
NVIDIA · Released March 19, 2026
Intelligence score
21
At a glance
- Intelligence
- 21
- Speed
- —
- Latency
- —
- Input price
- Not published
- Output price
- Not published
- Context window
- 2tokens
- Parameters
- 31.6billions
Summary
Comparison summary
Nemotron Cascade 2 30B A3B scores 21 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9).
Specifications
Technical specifications
| Model name | Nemotron Cascade 2 30B A3B |
|---|---|
| Provider | NVIDIA |
| Release date | March 19, 2026 |
| Intelligence score | 21 |
| Speed | — |
| Latency (TTFT) | — |
| Input price | Not published |
| Output price | Not published |
| Context window | 2 tokens |
| Parameters | 31.6B |
| Reasoning | Yes |
| Input modalities | text |
| Output modalities | — |
| Open weights | Yes |
| License | Nvidia Open Model License |
| API availability | — |
| Source | Model Intelligence legacy profile: nemotron-cascade-2-30b-a3b |
FAQ · 16
Frequently asked questions
Nemotron Cascade 2 30B A3B was released on March 19, 2026.
Nemotron Cascade 2 30B A3B was created by NVIDIA.
Nemotron Cascade 2 30B A3B scores 21 (estimated) on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9).
Yes, Nemotron Cascade 2 30B A3B is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
Nemotron Cascade 2 30B A3B supports text input.
Nemotron Cascade 2 30B A3B supports text output.
No, Nemotron Cascade 2 30B A3B does not support image input. It can only process text.
No, Nemotron Cascade 2 30B A3B is not multimodal. It only supports text input.
Nemotron Cascade 2 30B A3B has a context window of 1.0M tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Nemotron Cascade 2 30B A3B is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Nemotron Cascade 2 30B A3B has 31.6 billion parameters (3 billion active).
Nemotron Cascade 2 30B A3B is a Mixture of Experts (MoE) model with 31.6 billion total parameters, but only 3 billion active parameters are used during inference.
Nemotron Cascade 2 30B A3B is released under the Nvidia Open Model License license. This license allows commercial use.
Nemotron Cascade 2 30B A3B achieves a score of 21 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
Nemotron Cascade 2 30B A3B is an open weights model that can be self-hosted.
Nemotron Cascade 2 30B A3B is an open weights model that can be downloaded and self-hosted.