Compare / Gemma 3 4B vs Gemma 3n E2B
Gemma 3 4B vs Gemma 3n E2B
Updated November 2026
Verdict
Gemma 3 4B scores higher on the Small LM Index: 17.5 vs 14.3, although their likely ranges overlap, so the lead is not certain. On the 3 benchmarks both labs report, Gemma 3 4B wins 2 and Gemma 3n E2B wins 1. Both need about 4 GB of memory at 4-bit quantization. Only Gemma 3 4B has a hosted API on OpenRouter ($0.042 per million tokens); the other has to be self-hosted.
Pick Gemma 3 4B if…
- you want the strongest model overall
- you want a hosted API rather than running it yourself
- you work with long documents
- you want the more widely used model (more quantized builds and community support)
Pick Gemma 3n E2B if…
- you mostly write code
Specifications
| Gemma 3 4B | Gemma 3n E2B | |
|---|---|---|
| Small LM Index | 17.5 (likely 13–23) | 14.3 (likely 12–17) |
| Lab | Google · United States | Google · United States |
| Released | 2025-03-12 | 2025-06-26 |
| Parameters | 4.3B (dense) | 5.4B (dense) |
| Memory at 4-bit | ~3.5 GB | ~4 GB |
| Context window | 128K tokens | 32K tokens |
| Input | text, image | text, image, audio, video |
| Reasoning | No thinking | No thinking |
| License | Gemma | Gemma |
| API price / 1M tokens | $0.042 ($0.040 in, $0.050 out) | Self-host only |
| Hugging Face downloads (30 days) | 982K | 38.3K |
Benchmarks
Higher is better. Bold marks the winner where both labs report a score. ≈ marks the index model's estimate for a score that wasn't reported, ✓ an independent measurement and * a score measured by another lab.
| Benchmark | Gemma 3 4B | Gemma 3n E2B |
|---|---|---|
| MMLU-Pro Knowledge | 43.6 | 40.5 |
| GPQA Diamond Knowledge | 30.8 | 24.8 |
| AIME Math | ≈ 12 | 6.7 |
| LiveCodeBench Coding | 12.6 | 18.6 |
| MMLU Knowledge | — | 60.1 |
| Multilingual MMLU Multilingual | — | 55.1 |
Scores are mostly reported by each lab with its own settings, so small gaps are noise. See the methodology. Open in the interactive comparison to add more models.
Questions
Which is better, Gemma 3 4B or Gemma 3n E2B?
Gemma 3 4B scores higher on the Small LM Index: 17.5 vs 14.3, although their likely ranges overlap, so the lead is not certain. On the 3 benchmarks both labs report, Gemma 3 4B wins 2 and Gemma 3n E2B wins 1.
Which is cheaper to run?
Only Gemma 3 4B has a hosted API on OpenRouter ($0.042 per million tokens); the other has to be self-hosted. Both need about 4 GB of memory at 4-bit quantization.
Which has the longer context window?
Gemma 3 4B has the longer context window: 128K tokens against 32K.
Can I use Gemma 3 4B and Gemma 3n E2B commercially?
Both are released under the Gemma license. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.