Model comparison
Gemma 2B vs Gemma 4 31B IT
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 29.6 on the Noometry Index.
Last verified . 12 shared benchmarks.
GoogleSummary
- They share 12 benchmarks with published results for both. Gemma 2B scores higher in 0 categories and Gemma 4 31B IT in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 4 31B IT leads 60.5 to 24.0.
Side by side
| Gemma 2B | Gemma 4 31B IT | |
|---|---|---|
| Provider | ||
| Noometry Index | 29.6 | 43.5 |
| Released | 2024-02-21 | 2026-04-02 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 33K |
| Input $ / M tokens | — | $0.09 |
| Output $ / M tokens | — | $0.34 |
| Results tracked | 23 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 2B: 29.4 (#305), Gemma 4 31B IT: 42.3 (#108)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Coding | 1010 | 1459 |
| LMArena WebDev | — | 1366 |
| SciCode | — | 43.4% |
| WeirdML | — | 52.3% |
| ALE-Bench | — | 925.5 |
| HumanEval+ | 20.7% | — |
| MBPP+ | 34.1% | — |
Reasoning Gemma 4 31B IT leads
Gemma 2B: 18.8 (#275), Gemma 4 31B IT: 27.2 (#122)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Hard Prompts | 989 | 1448 |
| Epoch Capabilities Index | 94.2 | 142.74 |
| Kagi LLM Benchmark | — | 63.5% |
| NYT Connections (extended) | — | 70.6% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 5% |
| Thematic Generalization | — | 53% |
| DTBench | — | 82.7% |
| LMCA | — | 39.3% |
| Surface Evolver Bench | — | 30.6% |
| BIG-Bench Hard | 35.2% | — |
| HellaSwag | 71.4% | — |
| PIQA | 77.3% | — |
| WinoGrande | 65.4% | — |
Math Gemma 4 31B IT leads
Gemma 2B: 30.0 (#239), Gemma 4 31B IT: 43.2 (#81)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Math | 1009 | 1465 |
| OTIS Mock AIME 2024-2025 | — | 73.3% |
| GSM8K | 17.7% | — |
Knowledge Not comparable
Gemma 2B: —, Gemma 4 31B IT: 37.9 (#151)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| GPQA Diamond | — | 75.8% |
| SimpleQA Verified | — | 10.4% |
| Vectara Hallucination Rate | — | 7.4% |
| LMArena Expert | — | 1465 |
| ARC (AI2) Challenge | 42.1% | — |
| BoolQ | 69.4% | — |
| MMLU | 42.3% | — |
| TriviaQA | 53.2% | — |
Multimodal Not comparable
Gemma 2B: —, Gemma 4 31B IT: 41.6 (#34)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Vision | — | 1277 |
| LMArena Document | — | 1425 |
Multilingual Gemma 4 31B IT leads
Gemma 2B: 23.0 (#294), Gemma 4 31B IT: 53.8 (#57)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Non-English | 958 | 1431 |
| LMArena Chinese | 986 | 1476 |
| LMArena Russian | 937 | 1460 |
| LMArena French | — | 1435 |
| LMArena Spanish | — | 1444 |
Instruction Following Gemma 4 31B IT leads
Gemma 2B: 48.5 (#302), Gemma 4 31B IT: 75.5 (#61)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Instruction Following | 970 | 1433 |
Long Context Gemma 4 31B IT leads
Gemma 2B: 29.9 (#291), Gemma 4 31B IT: 44.2 (#71)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Longer Query | 981 | 1446 |
Writing & Preference Gemma 4 31B IT leads
Gemma 2B: 24.0 (#308), Gemma 4 31B IT: 60.5 (#96)
| Benchmark | Gemma 2B | Gemma 4 31B IT |
|---|---|---|
| LMArena Text | 1002 | 1443 |
| LMArena Creative Writing | 987 | 1415 |
| LMArena Multi-Turn | 945 | 1452 |
| EQ-Bench Creative Writing | — | 1368 |
| EQ-Bench 4 | — | 1120 |
Frequently asked questions
Is Gemma 2B better than Gemma 4 31B IT?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 29.6 on the Noometry Index.
Is Gemma 2B 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 2B and Gemma 4 31B IT share?
12 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Gemma 4 31B IT has 35.