Compare / Nanbeige4.2 3B vs Gemma 3 4B
Nanbeige4.2 3B vs Gemma 3 4B
Updated November 2026
Verdict
Nanbeige4.2 3B scores higher on the Small LM Index: 70.4 vs 17.5. On the 2 benchmarks both labs report, Nanbeige4.2 3B wins 2 and Gemma 3 4B wins 0. Both need about 3.8 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 Nanbeige4.2 3B if…
- you want the strongest model overall
- you mostly write code
- you work with long documents
- you need a permissive license (Apache-2.0)
Pick Gemma 3 4B if…
- you want a hosted API rather than running it yourself
- you need image input
- you want the more widely used model (more quantized builds and community support)
Specifications
| Nanbeige4.2 3B | Gemma 3 4B | |
|---|---|---|
| Small LM Index | 70.4 (likely 69–72) | 17.5 (likely 13–23) |
| Lab | Nanbeige (Kanzhun) · China | Google · United States |
| Released | 2026-07-21 | 2025-03-12 |
| Parameters | 4.2B (dense) | 4.3B (dense) |
| Memory at 4-bit | ~3.8 GB | ~3.5 GB |
| Context window | 262K tokens | 128K tokens |
| Input | text | text, image |
| Reasoning | Always thinks | No thinking |
| License | Apache-2.0 | Gemma |
| API price / 1M tokens | Self-host only | $0.042 ($0.040 in, $0.050 out) |
| Hugging Face downloads (30 days) | 120.3K | 982K |
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 | Nanbeige4.2 3B | Gemma 3 4B |
|---|---|---|
| MMLU-Pro Knowledge | ≈ 85 | 43.6 |
| GPQA Diamond Knowledge | 87.4 | 30.8 |
| Humanity's Last Exam Reasoning | 17.8 | ≈ 1 |
| LiveCodeBench Coding | 72.5 | 12.6 |
| SWE-bench Verified Coding | 63.6 | ≈ 5 |
| Terminal-Bench Agentic | 44.1 | — |
| HMMT Math | 82.8 | — |
| AA-LCR Long context | 58.7 | — |
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, Nanbeige4.2 3B or Gemma 3 4B?
Nanbeige4.2 3B scores higher on the Small LM Index: 70.4 vs 17.5. On the 2 benchmarks both labs report, Nanbeige4.2 3B wins 2 and Gemma 3 4B wins 0.
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 3.8 GB of memory at 4-bit quantization.
Which has the longer context window?
Nanbeige4.2 3B has the longer context window: 262K tokens against 128K.
Can I use Nanbeige4.2 3B and Gemma 3 4B commercially?
Nanbeige4.2 3B is released under Apache-2.0 and Gemma 3 4B under Gemma. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.