Google: Gemma 3n 2B (free)

Google: Gemma 3n 2B (free)

google · Released Jul 9, 2025 Professional
Intelligence #10 / 576
82.0 Our Score
Speed #188 / 271
61.3 tokens / sec
Input
— Not priced
Output
— Not priced
Context #432 / 577
8,192 tokens

Analysis Summary

Google's Gemma 3n 2B is a compact, on-device model from the Gemma 3n family, released mid-2025. Its benchmark profile reflects its size: reasoning, coding, and agentic scores are all at the low end of the field, with a very small 8K context window and text-only output.

For businesses, this model is not suited to client-facing content, coding workflows, or any task requiring reliable multi-step reasoning. Its value lies in edge deployment, low-latency local inference, or cost-constrained environments where a small footprint is the primary requirement.

The free tier makes it accessible for experimentation, but teams with real production workloads should look to larger models in the Gemma or Gemini families for meaningful output quality.

Assessed June 6, 2026

Editorial notes

Gemma 3n 2B is a lightweight Google model with very limited reasoning and coding capability, suited only to simple on-device or low-resource tasks where size matters more than performance.

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

Performance Profile

Intelligence1.2Technical0.3Value0Content2.4
Intelligence 1.2/10
Technical 0.3/10
Content 2.4/10
Value 0/10

How Google: Gemma 3n 2B (free) compares

Google: Gemma 3n 2B (free) ranks #377 of 378 AI models we track for overall intelligence, #299 of 315 for coding, #287 of 289 for agentic tasks. Its 8K-token context window is larger than 25% of the models we list. Google: Gemma 3n 2B (free) is currently free to use via OpenRouter.

About Google: Gemma 3n 2B (free)

Gemma 3n E2B IT is a multimodal, instruction-tuned model developed by Google DeepMind, designed to operate efficiently at an effective parameter size of 2B while leveraging a 6B architecture. Based..

2B Parameters

Performance Indices

Source: Artificial Analysis

4.8 Intelligence Index
2.2 Coding Index
0.8 Agentic Index
10.3 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 22.9% Graduate-level scientific reasoning
HLE 4% Humanity's Last Exam
MMLU Pro 37.8% Multi-task language understanding
MATH 500 69.1% Mathematical problem-solving
AIME 9% Competition mathematics
AIME 2025 10.3% Competition mathematics (2025)
SciCode 5.2% Scientific computing

Technical

LiveCodeBench 9.5% Live coding evaluation
TerminalBench Hard 0.8% Agentic terminal tasks

Content

IFBench 22% Instruction following

Benchmark data from Artificial Analysis and Hugging Face

How does Google: Gemma 3n 2B (free) stack up?

Compare side-by-side with other professional models.

Compare Models

Model Information

OpenRouter ID google/gemma-3n-e2b-it:free
Providergoogle
Release Date July 9, 2025
Context Length8,192 tokens
Max Completion2,048 tokens
Status Active

Frequently asked questions about Google: Gemma 3n 2B (free)

How much does Google: Gemma 3n 2B (free) cost?

Google: Gemma 3n 2B (free) is currently available for free via OpenRouter.

What is the context window of Google: Gemma 3n 2B (free)?

Google: Gemma 3n 2B (free) has a context window of 8,192 tokens (8K).

Is Google: Gemma 3n 2B (free) good for coding?

On our coding benchmark index, Google: Gemma 3n 2B (free) ranks #299 of 315 models, placing it in the broader range of the field for code generation and debugging.

Who created Google: Gemma 3n 2B (free)?

Google: Gemma 3n 2B (free) is developed by Google and was released on July 9, 2025.