Gemma 3 270M

Gemma 3 270M

Google · Released Aug 14, 2025 Professional
Intelligence #10 / 576
82.0 Our Score
AA Index #354 / 378
7.7 Artificial Analysis
Input
— Not priced
Output
— Not priced
Context
— Not reported

Analysis Summary

Gemma 3 270M is one of Google's smallest released models, with an intelligence index of 7.7 and an agentic index of 9.1. Its livecodebench score of 0.003 and MMLU Pro of 0.055 confirm limited capability across coding and general knowledge tasks. The math index of 2.3 and AIME score of 0.023 place it at the bottom of the benchmarked field.

For businesses, this model has no practical role in standard workflows. Its ifbench score of 0.121 and tau2 of 0.091 indicate it cannot reliably follow instructions or complete structured tasks. It may have a narrow application in on-device or edge inference scenarios where model size is the binding constraint and task complexity is minimal.

No pricing data is available. Given its capability profile, Gemma 3 270M is best treated as a research or experimentation artefact rather than a production tool. Teams with edge deployment needs should evaluate larger Gemma variants or purpose-built small models with stronger benchmark profiles.

Assessed June 6, 2026

Editorial notes

Gemma 3 270M from Google is a very small model with an intelligence index of 7.7 and near-zero coding and agentic capability, suitable only for the most constrained edge deployment scenarios.

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

Performance Profile

Intelligence1.1Technical0.9Value0Content1.3
Intelligence 1.1/10
Technical 0.9/10
Content 1.3/10
Value 0/10

How Gemma 3 270M compares

Gemma 3 270M ranks #354 of 378 AI models we track for overall intelligence, #314 of 315 for coding, #261 of 289 for agentic tasks. Gemma 3 270M is currently free to use via OpenRouter.

Performance Indices

Source: Artificial Analysis

7.7 Intelligence Index
9.1 Agentic Index
2.3 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 22.4% Graduate-level scientific reasoning
HLE 4.2% Humanity's Last Exam
MMLU Pro 5.5% Multi-task language understanding
AIME 2025 2.3% Competition mathematics (2025)

Technical

LiveCodeBench 0.3% Live coding evaluation
τ²-Bench 9.1% Conversational agent benchmark

Content

IFBench 12.1% Instruction following

Benchmark data from Artificial Analysis and Hugging Face

How does Gemma 3 270M stack up?

Compare side-by-side with other professional models.

Compare Models

Model Information

ProviderGoogle
Release Date August 14, 2025
Status Active

Frequently asked questions about Gemma 3 270M

How much does Gemma 3 270M cost?

Gemma 3 270M is currently available for free via OpenRouter.

Is Gemma 3 270M good for coding?

On our coding benchmark index, Gemma 3 270M ranks #314 of 315 models, placing it in the broader range of the field for code generation and debugging.

Who created Gemma 3 270M?

Gemma 3 270M is developed by Google and was released on August 14, 2025.