Gemma 3n E2B Instruct vs Gemma 4 E2B

Wins 0 of 2 benchmarks
Wins 1 of 2 benchmarks
Humanity's Last Exam Humanity's Last Exam: Very hard expert questions across many fields, written to stump today’s models
Gemma 3n E2B Instruct4.2
Gemma 4 E2B4.8
MMMU-Pro (vision) MMMU-Pro (vision): College-level questions that require understanding images: charts, diagrams, photos
Gemma 3n E2B Instruct—
Gemma 4 E2B44.6

Specs

SpecGemma 3n E2B InstructGemma 4 E2B
Input, per 1M—$0.08
Output, per 1M—$0.40
Context32K128K
WeightsOpenOpen
ReleasedJun 2025Apr 2026
Memory (4-bit)2.7 GB2.6 GB
LicenceGemma termsApache-2.0
Commercial useRestrictedAllowed
See images✓✓
Read PDFs——
Hear audio—✓
Watch video——
Call tools—✓
Force a tool call??
Return strict JSONPartialPartial
Search the web??
Run code??
Use a computer??

What breaks when you switch

Gemma 3n E2B Instruct → Gemma 4 E2B
Nothing known.
Gemma 4 E2B → Gemma 3n E2B Instruct
Context: 128K → 32K
Memory (4-bit): 2.6 GB → 2.7 GB