Model Comparison

Gemma 2 9B vs Gemma 3n E2B Instructed LiteRT (Preview)

Gemma 2 9B significantly outperforms across most benchmarks.

Want to compare interactively?Try the playground

Performance Benchmarks

Comparative analysis across standard metrics

12 benchmarks

Gemma 2 9B outperforms in 10 benchmarks (ARC-C, ARC-E, BoolQ, HellaSwag, MMLU, Natural Questions, PIQA, Social IQa, TriviaQA, Winogrande), while Gemma 3n E2B Instructed LiteRT (Preview) is better at 2 benchmarks (HumanEval, MBPP).

Gemma 2 9B significantly outperforms across most benchmarks.

Fri May 01 2026 • llm-stats.com

Arena Performance

Human preference votes

Pricing Analysis

Price comparison per million tokens

Cost data unavailable.

Lowest available price from all providers
Fri May 01 2026 • llm-stats.com
Google
Gemma 2 9B
Input tokens$0.00
Output tokens$0.00
Best providerUnknown Organization
Google
Gemma 3n E2B Instructed LiteRT (Preview)
Input tokens$0.00
Output tokens$0.00
Best providerUnknown Organization
Notice missing or incorrect data?Start an Issue→

Model Size

Parameter count comparison

7.3B diff

Gemma 2 9B has 7.3B more parameters than Gemma 3n E2B Instructed LiteRT (Preview), making it 383.8% larger.

Google
Gemma 2 9B
9.2Bparameters
Google
Gemma 3n E2B Instructed LiteRT (Preview)
1.9Bparameters
9.2B
Gemma 2 9B
1.9B
Gemma 3n E2B Instructed LiteRT (Preview)

Input Capabilities

Supported data types and modalities

Gemma 3n E2B Instructed LiteRT (Preview) supports multimodal inputs, whereas Gemma 2 9B does not.

Gemma 3n E2B Instructed LiteRT (Preview) can handle both text and other forms of data like images, making it suitable for multimodal applications.

Gemma 2 9B

Text
Images
Audio
Video

Gemma 3n E2B Instructed LiteRT (Preview)

Text
Images
Audio
Video

License

Usage and distribution terms

Both models are licensed under Gemma.

Both models share the same licensing terms, providing consistent usage rights.

Gemma 2 9B

Gemma

Open weights

Gemma 3n E2B Instructed LiteRT (Preview)

Gemma

Open weights

Release Timeline

When each model was launched

Gemma 2 9B was released on 2024-06-27, while Gemma 3n E2B Instructed LiteRT (Preview) was released on 2025-05-20.

Gemma 3n E2B Instructed LiteRT (Preview) is 11 months newer than Gemma 2 9B.

Gemma 2 9B

Jun 27, 2024

1.8 years ago

Gemma 3n E2B Instructed LiteRT (Preview)

May 20, 2025

11 months ago

10mo newer

Knowledge Cutoff

When training data ends

Gemma 3n E2B Instructed LiteRT (Preview) has a documented knowledge cutoff of 2024-06-01, while Gemma 2 9B's cutoff date is not specified.

We can confirm Gemma 3n E2B Instructed LiteRT (Preview)'s training data extends to 2024-06-01, but cannot make a direct comparison without Gemma 2 9B's cutoff date.

Gemma 2 9B

—

Gemma 3n E2B Instructed LiteRT (Preview)

Jun 2024

Outputs Comparison

Notice missing or incorrect data?Start an Issue discussion→

Key Takeaways

Google
Higher ARC-C score (68.4% vs 51.7%)
Higher ARC-E score (88.0% vs 75.8%)
Higher BoolQ score (84.2% vs 76.4%)
Higher HellaSwag score (81.9% vs 72.2%)
Higher MMLU score (71.3% vs 60.1%)
Higher Natural Questions score (29.2% vs 15.5%)
Higher PIQA score (81.7% vs 78.9%)
Higher Social IQa score (53.4% vs 48.8%)
Higher TriviaQA score (76.6% vs 60.8%)
Higher Winogrande score (80.6% vs 66.8%)
Google
Supports multimodal inputs
Higher HumanEval score (66.5% vs 40.2%)
Higher MBPP score (56.6% vs 52.4%)

Detailed Comparison

AI Model Comparison Table
Feature
Google
Gemma 2 9B
Google
Gemma 3n E2B Instructed LiteRT (Preview)

FAQ

Common questions about Gemma 2 9B vs Gemma 3n E2B Instructed LiteRT (Preview)

Gemma 2 9B significantly outperforms across most benchmarks. Gemma 2 9B is made by Google and Gemma 3n E2B Instructed LiteRT (Preview) is made by Google. The best choice depends on your use case — compare their benchmark scores, pricing, and capabilities above.
Gemma 2 9B scores ARC-E: 88.0%, BoolQ: 84.2%, HellaSwag: 81.9%, PIQA: 81.7%, Winogrande: 80.6%. Gemma 3n E2B Instructed LiteRT (Preview) scores PIQA: 78.9%, BoolQ: 76.4%, ARC-E: 75.8%, HellaSwag: 72.2%, Winogrande: 66.8%.
Key differences include multimodal support (no vs yes). See the full comparison above for benchmark-by-benchmark results.