Model comparison
Gemma 4 31B IT vs Qwen2-72B
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 29.5 on the Noometry Index.
Last verified . 17 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. Gemma 4 31B IT scores higher in 8 categories and Qwen2-72B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 4 31B IT leads 60.7 to 41.0.
- The biggest single-benchmark swing is WeirdML: 52.3% for Gemma 4 31B IT and 11.3% for Qwen2-72B.
Side by side
| Gemma 4 31B IT | Qwen2-72B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 43.2 | 29.5 |
| Released | 2026-04-02 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 33K | — |
| Input $ / M tokens | $0.09 | — |
| Output $ / M tokens | $0.34 | — |
| Results tracked | 35 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 4 31B IT leads
Gemma 4 31B IT: 41.9 (#98), Qwen2-72B: 28.7 (#303)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| WeirdML | 52.3% | 11.3% |
| LMArena Coding | 1459 | 1196 |
| LMArena WebDev | 1364 | — |
| SciCode | 43.4% | — |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
| ALE-Bench | 925.5 | — |
Agentic & Tool Use Not comparable
Gemma 4 31B IT: —, Qwen2-72B: 16.9 (#138)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 26.4 (#119), Qwen2-72B: 22.2 (#184)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1448 | 1191 |
| Epoch Capabilities Index | 142.74 | 125.27 |
| 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% | — |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 41.3 (#72), Qwen2-72B: 27.9 (#232)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| LMArena Math | 1464 | 1235 |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
| MATH Level 5 | — | 39.1% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#141), Qwen2-72B: 21.2 (#270)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 75.8% | 40.8% |
| LMArena Expert | 1465 | 1171 |
| SimpleQA Verified | 10.4% | — |
| Vectara Hallucination Rate | 7.4% | — |
| MMLU | — | 82.4% |
Multimodal Not comparable
Gemma 4 31B IT: 41.0 (#32), Qwen2-72B: —
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| LMArena Vision | 1278 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 54.0 (#52), Qwen2-72B: 36.0 (#239)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1431 | 1176 |
| LMArena Chinese | 1478 | 1240 |
| LMArena French | 1436 | 1169 |
| LMArena Russian | 1462 | 1169 |
| LMArena Spanish | 1443 | 1168 |
| LMArena German | — | 1151 |
| LMArena Japanese | — | 1111 |
| LMArena Korean | — | 1083 |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.6 (#58), Qwen2-72B: 62.0 (#236)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1181 |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#67), Qwen2-72B: 36.1 (#230)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1446 | 1193 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.7 (#92), Qwen2-72B: 41.0 (#236)
| Benchmark | Gemma 4 31B IT | Qwen2-72B |
|---|---|---|
| LMArena Text | 1443 | 1203 |
| LMArena Creative Writing | 1416 | 1181 |
| LMArena Multi-Turn | 1452 | 1195 |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
Frequently asked questions
Is Gemma 4 31B IT better than Qwen2-72B?
Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 29.5 on the Noometry Index.
Is Gemma 4 31B IT or Qwen2-72B better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 41.9 versus 28.7 in the Noometry coding category.
How many benchmarks do Gemma 4 31B IT and Qwen2-72B share?
17 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Qwen2-72B has 26.