Gemma 3n E2B Instruct
Intelligence
4.8
#522/523
Coding
2.2
#408/429
Math
10.3
#232/265
Speed
50 tok/s
TTFT: 399.00s
Pricing
—
Gemma 3n E2B Instruct is Google’s model designed for advanced mathematical computations and coding tasks. It processes at 50.173 tokens per second, targeting professional users who require high-level analytical capabilities.
When to Use Gemma 3n E2B Instruct
✓ Best For
- Complex mathematical problem solving.
- Data analysis and interpretation.
- Advanced coding assistance.
✗ Not Ideal For
- Basic coding tasks.
- General conversational AI applications.
How Gemma 3n E2B Instruct Compares
Intelligence Index · Higher is better
GoogleIBMCohere
Benchmark Profile
Coding Index
Liquid AIAlibabaGoogleSwiss AI InitiativeOpenBMB
Math Index
AI21 LabsAlibabaGoogleLiquid AINVIDIA
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (24)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 4.8 | |
| Coding Index | 2.2 | |
| Math Index | 10.3 | |
| MMLU-Pro | 378% | |
| GPQA | 229% | |
| LiveCodeBench | 95% | |
| HLE | 4% | |
| SciCode | 5.2% | |
| IFBench | 22% | |
| TerminalBench Hard | 0.8% | |
| AIME | 9% | |
| AIME 2025 | 10.3% | |
| MATH 500 | 69.1% | |
| ARC-C | 51.7% | |
| ARC-E | 75.8% | |
| BIG-Bench Hard | 44.3% | |
| BoolQ | 76.4% | |
| DROP | 53.9% | |
| HellaSwag | 72.2% | |
| Natural Questions | 15.5% | |
| PIQA | 78.9% | |
| Social IQa | 48.8% | |
| TriviaQA | 60.8% | |
| Winogrande | 66.8% |
Data: Artificial Analysis · Updated: April 10, 2026
Frequently Asked Questions (15)
When was Gemma 3n E2B Instruct released?
Gemma 3n E2B Instruct was released on June 26, 2025.
Who created Gemma 3n E2B Instruct?
Gemma 3n E2B Instruct was created by Google.
How intelligent is Gemma 3n E2B Instruct?
Gemma 3n E2B Instruct scores 5 on the Artificial Analysis Intelligence Index, placing it at the lower end among other open weight non-reasoning models of similar size (median: 11).
How verbose is Gemma 3n E2B Instruct?
When evaluated on the Intelligence Index, Gemma 3n E2B Instruct generated 4.1M output tokens, which is very competitive compared to other open weight non-reasoning models of similar size (median: 8.5M).
Is Gemma 3n E2B Instruct a reasoning model?
No, Gemma 3n E2B Instruct is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
What input modalities does Gemma 3n E2B Instruct support?
Gemma 3n E2B Instruct supports text and image input.
What output modalities does Gemma 3n E2B Instruct support?
Gemma 3n E2B Instruct supports text output.
Can Gemma 3n E2B Instruct process images?
Yes, Gemma 3n E2B Instruct supports image input and can analyze, describe, and answer questions about images.
Is Gemma 3n E2B Instruct multimodal?
Yes, Gemma 3n E2B Instruct is multimodal. It can process text and image input and generate text output.
What is the context window of Gemma 3n E2B Instruct?
Gemma 3n E2B Instruct has a context window of 32k tokens. This determines how much text and conversation history the model can process in a single request.
Is Gemma 3n E2B Instruct open source?
Yes, Gemma 3n E2B Instruct is open weights. The model weights are publicly available and can be downloaded for self-hosting.
How many parameters does Gemma 3n E2B Instruct have?
Gemma 3n E2B Instruct has 5.98 billion parameters (2 billion active).
What are the active parameters of Gemma 3n E2B Instruct?
Gemma 3n E2B Instruct is a Mixture of Experts (MoE) model with 5.98 billion total parameters, but only 2 billion active parameters are used during inference.
What is the license for Gemma 3n E2B Instruct?
Gemma 3n E2B Instruct is released under the Gemma license license. This license allows commercial use.
How does Gemma 3n E2B Instruct perform on benchmarks?
Gemma 3n E2B Instruct achieves a score of 5 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
