Nemotron-Cascade Models by NVIDIA AI
NVIDIA AI releases · 27 in the last 12 months · this family litChangelog →
1 model2026Up to 1M ctx
Last refreshed 2026-05-14. Next refresh: weekly.
Details
ResearcherNVIDIA AI
LicenseLlama 3 Community
Commercial useRestricted
Models1
Released2026
Max context1M
Links
WebsiteAbout
Cascade MoE reasoning models with superior performance on math and code tasks
Decision facts
- Best fit
- codingmath-heavy prompts
- Capability starting point
- Nemotron-Cascade-2-30B-A3B with 1M context
- Lowest tracked input
- Not tracked
- Closest related family
- NemoGuard
Current Variants
Use-when guidance is based on each model's tracked capabilities, context window, release date, and replacement status.
1 in view
Nemotron-Cascade-2-30B-A3BCurrent
Use when the workload needs 1M context and 30B parameters.
2026-031M context30B parameters
| Model | Use when | Released | Signals | Status |
|---|---|---|---|---|
| Nemotron-Cascade-2-30B-A3B | Use when the workload needs 1M context and 30B parameters. | 2026-03 | 1M context30B parameters | Current |
Release Timeline
1 release group2026-03
1 current
Nemotron-Cascade-2-30B-A3B
Current1M context30B parameters
Specifications(1 models)
| Model | Released | Context | Parameters |
|---|---|---|---|
| Nemotron-Cascade-2-30B-A3B | 2026-03 | 1M | 30B |
