Amazon Nova Premier vs Google Gemma 3n E4B Instructed

Detailed comparison for LLMs

AmazonGoogle

On Guardion's LLM vulnerability Benchmark, Amazon Nova Premier is the more secure of the two: Nova Premier scores 20.0% and Gemma 3n E4B Instructed scores 49.0% on attack success rate (ASR) (lower is better). One or both scores are estimated from public safety evaluations pending a Guardion benchmark run.

Head-to-Head Overview

Nova Premier is the overall winner in this comparison!

Attack Success Rate (lower is safer)

ASR for Amazon Nova Premier vs Google Gemma 3n E4B Instructed. Green marks the safer model on each metric. Only the overall score is available for estimated models.

Overall (ASR)

Nova Premier
20.0%
Gemma 3n E4B Instructed
49.0%

TAP Attack Method (ASR)

Nova Premier
Gemma 3n E4B Instructed
100.0%

Crescendo Attack Method (ASR)

Nova Premier
Gemma 3n E4B Instructed
100.0%

Zero-Shot (ASR)

Nova Premier
Gemma 3n E4B Instructed
100.0%

Key Highlights

  • Amazon Nova Premier has a lower Overall (ASR).

Security Profile

Outward is better on every axis.

OverallTAPCrescendoZero-Shot
Nova Premier
Gemma 3n E4B Instructed
Full security profile
Amazon Nova Premier
Full security profile
Google Gemma 3n E4B Instructed

Frequently asked questions

Is Amazon Nova Premier or Google Gemma 3n E4B Instructed more secure?

On Guardion's LLM vulnerability Benchmark, Amazon Nova Premier is the more secure of the two: Nova Premier scores 20.0% and Gemma 3n E4B Instructed scores 49.0% on attack success rate (ASR) (lower is better). One or both scores are estimated from public safety evaluations pending a Guardion benchmark run.

What is the attack success rate (ASR) of Nova Premier vs Gemma 3n E4B Instructed?

Nova Premier has a 20.0% ASR and Gemma 3n E4B Instructed has a 49.0% ASR — the share of adversarial prompts that succeed across zero-shot, TAP, and Crescendo attacks. Lower is safer.

How were Nova Premier and Gemma 3n E4B Instructed tested?

Both were red-teamed with the HarmBench framework across zero-shot, TAP (Tree of Attacks with Pruning), and Crescendo multi-turn attacks, scored by Attack Success Rate.

Related Comparisons