Gemma 4 E2B Benchmark Details

Gemma 4 E2B currently shows benchmark results led by MMLU-Pro (123 / 176, score 60), LiveCodeBench (180 / 251, score 44), IF Bench (230 / 282, score 38). This page also compares it with 2 competitor models and 1 predecessor or same-series models, including performance and pricing views when available.

Benchmark Results

Gemma 4 E2B

Benchmark Results

General Knowledge

3 evaluations

Benchmark / mode

Score

Rank/total

MMLU-Pro

Thinking Mode

123 / 176

HLE

Standard Mode

4.70

492 / 565

HLE

Thinking Mode

4.80

490 / 565

Other

2 evaluations

Benchmark / mode

Score

Rank/total

GPQA Diamond

Standard Mode

40.50

432 / 463

GPQA Diamond

Thinking Mode

43.30

423 / 463

Coding and Software Engineer

2 evaluations

Benchmark / mode

Score

Rank/total

CodeForces

Thinking Mode

633

21 / 21

LiveCodeBench

Thinking Mode

180 / 251

Agent Level Benchmark

5 evaluations

Benchmark / mode

Score

Rank/total

τ²-Bench

Thinking Mode

24.50

44 / 44

τ²-Bench - Telecom

Standard ModeTools

22.20

238 / 264

τ²-Bench - Telecom

Thinking ModeTools

20.80

246 / 264

Terminal Bench Hard

Standard ModeTools

2.30

229 / 244

Terminal Bench Hard

Thinking ModeTools

225 / 244

Instruction Following

2 evaluations

Benchmark / mode

Score

Rank/total

IF Bench

Standard Mode

33.60

250 / 282

IF Bench

Thinking Mode

230 / 282

Long Context

1 evaluations

Benchmark / mode

Score

Rank/total

AA-LCR

Standard Mode

16.30

170 / 171

Math and Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

AIME 2026

Thinking Mode

37.50

30 / 30

AI Agent - Tool Usage

1 evaluations

Benchmark / mode

Score

Rank/total

Terminal-Bench 2.1

Thinking ModeTools

0.40

193 / 194

Multimodal Understanding

4 evaluations

Benchmark / mode

Score

Rank/total

MathVision

Thinking Mode

52.40

12 / 12

MMMU-Pro

Standard Mode

41.80

220 / 229

MMMU-Pro

unknown

44.20

214 / 229

MMMU-Pro

Thinking Mode

44.20

214 / 229

Other

1 evaluations

Benchmark / mode

Score

Rank/total

GDM-MRCR v2 (8-needle, 128K)

Thinking Mode

19.10

8 / 8

Compare with other models

Competitor Comparison

Benchmark scores for Gemma 4 E2B compared against top models in its class

Gemma 4 E2BMiniCPM5-1B MiniCPM-V 4.6

Benchmark categories:

The chart shows each model’s highest score per benchmark within the current filter. Out-of-100 benchmarks use raw heights; out-of-range benchmarks are scaled within that benchmark while labels keep the original scores.

6 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.

Benchmark	Gemma 4 E2BCurrent	MiniCPM5-1B
HLE Accuracy 综合评估	4.80Thinking Enabled	6.50Thinking Enabled
MMLU-Pro Accuracy 综合评估	60.00Thinking Enabled	48.85Thinking Enabled
GPQA Diamond Accuracy 科学与综合推理	43.30Thinking Enabled	26.90Standard Mode
τ²-Bench - Telecom Accuracy Agent能力评测	22.20Standard Mode ｜ Tools	82.50Standard Mode ｜ Tools
IF Bench Accuracy 指令跟随	38.00Thinking Enabled	49.30Thinking Enabled
AA-LCR Accuracy 长上下文能力	16.30Standard Mode	5.00Standard Mode

Standard API Pricing: Gemma 4 E2B vs. Peer Models

Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.

Source: DataLearnerAI. Standard text prices shown here use the default supplier.

Comparable standard text pricing is not available for these models.

Version History

How each version of the Gemma 4 E2B series stacks up on benchmark tests

Gemma 4 E2BGemma-3n-E2B

No benchmark data matches the selected filters.

Standard API Pricing Across the Gemma 4 E2B Series

Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.

Source: DataLearnerAI. Standard text prices shown here use the default supplier. · USD / 1M tokens

Model	Supplier	Standard input	Standard output	Base price applies to
Gemma-3n-E2B	Google DeepMind	$0 / 1M tokens	$0 / 1M tokens	—