Nemotron-3-Super-120B-A12B vs Nemotron-Cascade 2

Data-driven comparison powered by the gentic.news knowledge graph

Nemotron-3-Super-120B-A12B:↑ rising

Nemotron-Cascade 2:→ stable

competes with (3 sources)

Nemotron-3-Super-120B-A12B

ai model

METRIC

Nemotron-Cascade 2

ai model

Total Mentions

Last 30 Days

Last 7 Days

↑ rising

Momentum

→ stable

Positive (+0.50)

Sentiment (30d)

Positive (+0.50)

Mar 22, 2026

First Covered

Mar 20, 2026

Ecosystem

Nemotron-3-Super-120B-A12B

No mapped relationships

Nemotron-Cascade 2

competes withQwen 3.5 Medium11 sources

usesMixture-of-Experts9 sources

competes withNemotron-3-Super-120B-A12B3 sources

usesLiveCodeBench v61 sources

usesInternational Olympiad in Informatics 20251 sources

usesHugging Face Hub1 sources

Nemotron-3-Super-120B-A12B

NVIDIA's Nemotron-3-Super-120B-A12B is a 120-billion parameter open hybrid MoE model that activates only 12B parameters for efficiency, combining Mamba and Transformer layers for advanced reasoning in agentic systems.

Nemotron-Cascade 2

NVIDIA's Nemotron-Cascade 2 is an open 30B parameter Mixture-of-Experts model with only 3B active parameters, achieving top-tier mathematical reasoning and coding performance.

Recent Events

Nemotron-3-Super-120B-A12B

No timeline events

Nemotron-Cascade 2

2026-03-22

Achieved Gold Medal-level performance on 2025 International Mathematical Olympiad, International Olympiad in Informatics, and ICPC World Finals

2026-03-20

Achieved 'gold medal performance' on IMO 2025 and IOI 2025 benchmarks

2025-01-01

Achieved Gold Medal-level performance on 2025 International Mathematical Olympiad, International Olympiad in Informatics, and ICPC World Finals

Related Comparisons

Nemotron-Cascade 2 vs Qwen 3.5 Medium

11 evidence sources

Nemotron-3-Super-120B-A12B Profile|Nemotron-Cascade 2 Profile|Knowledge Graph