Model comparison
Gemma 2 9B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 25.9 on the Noometry Index.
Last verified . 24 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Gemma 2 9B scores higher in 1 category and Qwen2.5-Coder-32B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Coder-32B leads 33.4 to 9.7.
- The biggest single-benchmark swing is LiveBench Coding: 22.5% for Gemma 2 9B and 56.9% for Qwen2.5-Coder-32B.
Side by side
| Gemma 2 9B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 25.9 | 33.4 |
| Released | 2024-06-24 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 35 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2 9B leads
Gemma 2 9B: 29.4 (#304), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| BigCodeBench Instruct | 34.7% | 49% |
| LiveBench Coding | 22.5% | 56.9% |
| LMArena Coding | 1173 | 1276 |
| BigCodeBench Complete | 40.6% | 58% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Qwen2.5-Coder-32B leads
Gemma 2 9B: 15.9 (#309), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Reasoning | 15.2% | 42.1% |
| LMArena Hard Prompts | 1171 | 1251 |
| LiveBench Data Analysis | 36.4% | 49.9% |
| Epoch Capabilities Index | 119.83 | 119.49 |
| LiveBench | 28.7% | 46.2% |
| HellaSwag | — | 83% |
| PIQA | 83.7% | — |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Gemma 2 9B: 9.9 (#318), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Math | 19.8% | 46.6% |
| LMArena Math | 1183 | 1251 |
| GSM8K | 84.9% | 93% |
| OTIS Mock AIME 2024-2025 | 0.6% | — |
| MATH Level 5 | 21% | — |
Knowledge Qwen2.5-Coder-32B leads
Gemma 2 9B: 9.7 (#305), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1147 | 1221 |
| MMLU | 72.1% | 79.1% |
| GPQA Diamond | 27.5% | — |
| ARC (AI2) Challenge | — | 70.5% |
| BoolQ | 85.7% | — |
Multilingual Qwen2.5-Coder-32B leads
Gemma 2 9B: 36.6 (#238), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1188 | 1205 |
| LMArena Chinese | 1185 | 1222 |
| LMArena Russian | 1200 | 1228 |
| LMArena French | 1190 | — |
| LMArena German | 1186 | — |
| LMArena Japanese | 1144 | — |
| LMArena Korean | 1137 | — |
| LMArena Spanish | 1200 | — |
Instruction Following Qwen2.5-Coder-32B leads
Gemma 2 9B: 57.6 (#269), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Instruction Following | 52.6% | 58.7% |
| LMArena Instruction Following | 1178 | 1223 |
Long Context Qwen2.5-Coder-32B leads
Gemma 2 9B: 36.3 (#233), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1197 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Gemma 2 9B: 32.1 (#281), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Gemma 2 9B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1207 | 1230 |
| LMArena Creative Writing | 1206 | 1174 |
| LMArena Multi-Turn | 1193 | 1222 |
| LiveBench Language | 25.5% | 23.3% |
| EQ-Bench Creative Writing | 841 | — |
Frequently asked questions
Is Gemma 2 9B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 25.9 on the Noometry Index.
Is Gemma 2 9B or Qwen2.5-Coder-32B better for coding?
Gemma 2 9B scores higher on coding benchmarks: 29.4 versus 22.6 in the Noometry coding category.
How many benchmarks do Gemma 2 9B and Qwen2.5-Coder-32B share?
24 benchmarks have published results for both models. Gemma 2 9B has 35 scored results on Noometry and Qwen2.5-Coder-32B has 31.