Nemotron Cascade 2 30B A3B

Nemotron Cascade 2 30B A3B

NVIDIA · Released Mar 19, 2026 Efficient
Intelligence #235 / 557
36.8 Our Score
AA Index #109 / 365
28.4 Artificial Analysis
Input
Not priced
Output
Not priced
Context
Not reported

Analysis Summary

Nemotron Cascade 2 30B A3B sits in the Efficient tier on our leaderboard, ranked #235 of 557 published models on overall intelligence. At $0.000 input and $0.000 output per 1M tokens, it is among the most expensive on the market.

Editorial notes

Nemotron Cascade 2 30B A3B shows strong instruction-following and reasonable agentic capability for a mid-size model, with a notable ifbench score, though no pricing data limits value assessment.

Assessed May 14, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence5.1Technical4.4Value0Content5.5
Intelligence 5.1/10
Technical 4.4/10
Content 5.5/10
Value 0/10

Performance Indices

Source: Artificial Analysis

28.4 Intelligence Index
25.8 Coding Index
37.2 Agentic Index

Benchmark Scores

Intelligence

GPQA Diamond 75.8% Graduate-level scientific reasoning
HLE 11.4% Humanity's Last Exam
SciCode 34.8% Scientific computing

Technical

TerminalBench Hard 21.2% Agentic terminal tasks
τ²-Bench 53.2% Conversational agent benchmark

Content

IFBench 80.4% Instruction following
LCR 34% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does Nemotron Cascade 2 30B A3B stack up?

Compare side-by-side with other efficient models.

Compare Models

Model Information

ProviderNVIDIA
Release Date March 19, 2026
Status Active

Leaderboard Categories