Model comparison
Gemma 2 9B vs Gemma 4 31B IT
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 25.9 on the Noometry Index.
Last verified . 18 shared benchmarks.
GoogleSummary
- They share 18 benchmarks with published results for both. Gemma 2 9B scores higher in 0 categories and Gemma 4 31B IT in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemma 4 31B IT leads 43.2 to 9.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 0.6% for Gemma 2 9B and 73.3% for Gemma 4 31B IT.
Side by side
| Gemma 2 9B | Gemma 4 31B IT | |
|---|---|---|
| Provider | ||
| Noometry Index | 25.9 | 43.5 |
| Released | 2024-06-24 | 2026-04-02 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 33K |
| Input $ / M tokens | — | $0.09 |
| Output $ / M tokens | — | $0.34 |
| Results tracked | 35 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 2 9B: 29.4 (#304), Gemma 4 31B IT: 42.3 (#108)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| LMArena Coding | 1173 | 1459 |
| LMArena WebDev | — | 1366 |
| SciCode | — | 43.4% |
| WeirdML | — | 52.3% |
| BigCodeBench Instruct | 34.7% | — |
| LiveBench Coding | 22.5% | — |
| BigCodeBench Complete | 40.6% | — |
| ALE-Bench | — | 925.5 |
Reasoning Gemma 4 31B IT leads
Gemma 2 9B: 15.9 (#309), Gemma 4 31B IT: 27.2 (#122)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| LMArena Hard Prompts | 1171 | 1448 |
| Epoch Capabilities Index | 119.83 | 142.74 |
| Kagi LLM Benchmark | — | 63.5% |
| NYT Connections (extended) | — | 70.6% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 5% |
| Thematic Generalization | — | 53% |
| LiveBench Reasoning | 15.2% | — |
| DTBench | — | 82.7% |
| LiveBench Data Analysis | 36.4% | — |
| LMCA | — | 39.3% |
| Surface Evolver Bench | — | 30.6% |
| LiveBench | 28.7% | — |
| PIQA | 83.7% | — |
Math Gemma 4 31B IT leads
Gemma 2 9B: 9.9 (#318), Gemma 4 31B IT: 43.2 (#81)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.6% | 73.3% |
| LMArena Math | 1183 | 1465 |
| LiveBench Math | 19.8% | — |
| MATH Level 5 | 21% | — |
| GSM8K | 84.9% | — |
Knowledge Gemma 4 31B IT leads
Gemma 2 9B: 9.7 (#305), Gemma 4 31B IT: 37.9 (#151)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| GPQA Diamond | 27.5% | 75.8% |
| LMArena Expert | 1147 | 1465 |
| SimpleQA Verified | — | 10.4% |
| Vectara Hallucination Rate | — | 7.4% |
| BoolQ | 85.7% | — |
| MMLU | 72.1% | — |
Multimodal Not comparable
Gemma 2 9B: —, Gemma 4 31B IT: 41.6 (#34)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| LMArena Vision | — | 1277 |
| LMArena Document | — | 1425 |
Multilingual Gemma 4 31B IT leads
Gemma 2 9B: 36.6 (#238), Gemma 4 31B IT: 53.8 (#57)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| LMArena Non-English | 1188 | 1431 |
| LMArena Chinese | 1185 | 1476 |
| LMArena French | 1190 | 1435 |
| LMArena Russian | 1200 | 1460 |
| LMArena Spanish | 1200 | 1444 |
| LMArena German | 1186 | — |
| LMArena Japanese | 1144 | — |
| LMArena Korean | 1137 | — |
Instruction Following Gemma 4 31B IT leads
Gemma 2 9B: 57.6 (#269), Gemma 4 31B IT: 75.5 (#61)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| LMArena Instruction Following | 1178 | 1433 |
| LiveBench Instruction Following | 52.6% | — |
Long Context Gemma 4 31B IT leads
Gemma 2 9B: 36.3 (#233), Gemma 4 31B IT: 44.2 (#71)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| LMArena Longer Query | 1197 | 1446 |
Writing & Preference Gemma 4 31B IT leads
Gemma 2 9B: 32.1 (#281), Gemma 4 31B IT: 60.5 (#96)
| Benchmark | Gemma 2 9B | Gemma 4 31B IT |
|---|---|---|
| LMArena Text | 1207 | 1443 |
| LMArena Creative Writing | 1206 | 1415 |
| EQ-Bench Creative Writing | 841 | 1368 |
| LMArena Multi-Turn | 1193 | 1452 |
| EQ-Bench 4 | — | 1120 |
| LiveBench Language | 25.5% | — |
Frequently asked questions
Is Gemma 2 9B better than Gemma 4 31B IT?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 25.9 on the Noometry Index.
Is Gemma 2 9B or Gemma 4 31B IT better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2 9B and Gemma 4 31B IT share?
18 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and Gemma 4 31B IT has 35.