Model comparison
Gemma 4 31B IT vs Qwen2.5-Coder-32B
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 33.4 on the Noometry Index.
Last verified . 13 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Gemma 4 31B IT scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Gemma 4 31B IT leads 42.3 to 22.6.
- Gemma 4 31B IT is cheaper at $0.09 / $0.34 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Gemma 4 31B IT accepts more context: 262K tokens versus 33K.
Side by side
| Gemma 4 31B IT | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 43.5 | 33.4 |
| Released | 2026-04-02 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 262K | 33K |
| Max output | 33K | 29K |
| Input $ / M tokens | $0.09 | $0.66 |
| Output $ / M tokens | $0.34 | $1 |
| Results tracked | 35 | 31 |
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: 42.3 (#108), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1459 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LMArena WebDev | 1366 | — |
| SciCode | 43.4% | — |
| WeirdML | 52.3% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 925.5 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Gemma 4 31B IT leads
Gemma 4 31B IT: 27.2 (#122), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1448 | 1251 |
| Epoch Capabilities Index | 142.74 | 119.49 |
| Kagi LLM Benchmark | 63.5% | — |
| NYT Connections (extended) | 70.6% | — |
| CritPt | 1.4% | — |
| Chess Puzzles | 5% | — |
| Thematic Generalization | 53% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 82.7% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 39.3% | — |
| Surface Evolver Bench | 30.6% | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Gemma 4 31B IT leads
Gemma 4 31B IT: 43.2 (#81), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1465 | 1251 |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Gemma 4 31B IT leads
Gemma 4 31B IT: 37.9 (#151), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1465 | 1221 |
| GPQA Diamond | 75.8% | — |
| SimpleQA Verified | 10.4% | — |
| Vectara Hallucination Rate | 7.4% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multimodal Not comparable
Gemma 4 31B IT: 41.6 (#34), Qwen2.5-Coder-32B: —
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Vision | 1277 | — |
| LMArena Document | 1425 | — |
Multilingual Gemma 4 31B IT leads
Gemma 4 31B IT: 53.8 (#57), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1431 | 1205 |
| LMArena Chinese | 1476 | 1222 |
| LMArena Russian | 1460 | 1228 |
| LMArena French | 1435 | — |
| LMArena Spanish | 1444 | — |
Instruction Following Gemma 4 31B IT leads
Gemma 4 31B IT: 75.5 (#61), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1433 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Gemma 4 31B IT leads
Gemma 4 31B IT: 44.2 (#71), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1446 | 1251 |
Writing & Preference Gemma 4 31B IT leads
Gemma 4 31B IT: 60.5 (#96), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Gemma 4 31B IT | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1443 | 1230 |
| LMArena Creative Writing | 1415 | 1174 |
| LMArena Multi-Turn | 1452 | 1222 |
| EQ-Bench Creative Writing | 1368 | — |
| EQ-Bench 4 | 1120 | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Gemma 4 31B IT better than Qwen2.5-Coder-32B?
Gemma 4 31B IT is the stronger model overall, scoring 43.5 to 33.4 on the Noometry Index.
Which is cheaper, Gemma 4 31B IT or Qwen2.5-Coder-32B?
Gemma 4 31B IT is cheaper. It lists at $0.09 per million input tokens and $0.34 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Gemma 4 31B IT or Qwen2.5-Coder-32B better for coding?
Gemma 4 31B IT scores higher on coding benchmarks: 42.3 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Gemma 4 31B IT does, with 262K tokens against 33K.
How many benchmarks do Gemma 4 31B IT and Qwen2.5-Coder-32B share?
13 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Qwen2.5-Coder-32B has 31.