Compare / Gemma 4 31B vs Granite 4.2 30B
Gemma 4 31B vs Granite 4.2 30B
Updated November 2026
Verdict
Gemma 4 31B scores higher on the Small LM Index: 70.1 vs 67.0. On the 5 benchmarks both labs report, Gemma 4 31B wins 4 and Granite 4.2 30B wins 0. Both need about 21.9 GB of memory at 4-bit quantization. Only Gemma 4 31B has a hosted API on OpenRouter ($0.16 per million tokens); the other has to be self-hosted.
Pick Gemma 4 31B if…
- you want the strongest model overall
- you mostly write code
- you build agents that call tools
- you want a hosted API rather than running it yourself
- you work with long documents
- you need image input
- you want the more widely used model (more quantized builds and community support)
Pick Granite 4.2 30B if…
- you need math and reasoning
Specifications
| Gemma 4 31B | Granite 4.2 30B | |
|---|---|---|
| Small LM Index | 70.1 (likely 69–72) | 67.0 (likely 66–69) |
| Lab | Google · United States | IBM (Granite) · United States |
| Released | 2026-04-02 | 2026-08-07 |
| Parameters | 30.7B (dense) | 30B (dense) |
| Memory at 4-bit | ~21.9 GB | ~21.4 GB |
| Context window | 256K tokens | 131K tokens |
| Input | text, image | text |
| Reasoning | Hybrid (thinking on/off) | No thinking |
| License | Apache-2.0 | Apache-2.0 |
| API price / 1M tokens | $0.16 ($0.10 in, $0.34 out) | Self-host only |
| Hugging Face downloads (30 days) | 8M | 32K |
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 4 31B | Granite 4.2 30B |
|---|---|---|
| MMLU-Pro Knowledge | 85.2 | 77.6 |
| GPQA Diamond Knowledge | 84.3 | 66.4 |
| Humanity's Last Exam Reasoning | 19.5 | ≈ 15 |
| AIME Math | 89.2 | 89.2 |
| LiveCodeBench Coding | 80.0 | 75.8 |
| SWE-bench Verified Coding | 52.0* | ≈ 58 |
| Terminal-Bench Agentic | 42.9* | 29.2 |
| τ²-Bench Agentic | 76.9 | — |
| IFBench Instruction following | — | 77.2 |
| MMMU-Pro Vision | 76.9 | — |
| BFCL Agentic | — | 61.4 |
| HMMT Math | — | 89.2 |
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 4 31B or Granite 4.2 30B?
Gemma 4 31B scores higher on the Small LM Index: 70.1 vs 67.0. On the 5 benchmarks both labs report, Gemma 4 31B wins 4 and Granite 4.2 30B wins 0.
Which is cheaper to run?
Only Gemma 4 31B has a hosted API on OpenRouter ($0.16 per million tokens); the other has to be self-hosted. Both need about 21.9 GB of memory at 4-bit quantization.
Which has the longer context window?
Gemma 4 31B has the longer context window: 256K tokens against 131K.
Can I use Gemma 4 31B and Granite 4.2 30B commercially?
Both are released under the Apache-2.0 license. Apache 2.0 and MIT allow commercial use; other licenses can add conditions, so read the license before shipping a product.