Model comparison
Gemma 2B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.6 on the Noometry Index.
Last verified . 19 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 19 benchmarks with published results for both. Gemma 2B scores higher in 1 category and Qwen2.5-Coder-32B in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5-Coder-32B leads 41.6 to 24.0.
Side by side
| Gemma 2B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 29.6 | 33.4 |
| Released | 2024-02-21 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 23 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2B leads
Gemma 2B: 29.4 (#305), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1010 | 1276 |
| HumanEval+ | 20.7% | 87.2% |
| MBPP+ | 34.1% | 77% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
Reasoning Qwen2.5-Coder-32B leads
Gemma 2B: 18.8 (#275), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 989 | 1251 |
| Epoch Capabilities Index | 94.2 | 119.49 |
| HellaSwag | 71.4% | 83% |
| WinoGrande | 65.4% | 80.8% |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| BIG-Bench Hard | 35.2% | — |
| LiveBench | — | 46.2% |
| PIQA | 77.3% | — |
Math Qwen2.5-Coder-32B leads
Gemma 2B: 30.0 (#239), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1009 | 1251 |
| GSM8K | 17.7% | 93% |
| LiveBench Math | — | 46.6% |
Knowledge Not comparable
Gemma 2B: —, Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| ARC (AI2) Challenge | 42.1% | 70.5% |
| MMLU | 42.3% | 79.1% |
| LMArena Expert | — | 1221 |
| BoolQ | 69.4% | — |
| TriviaQA | 53.2% | — |
Multilingual Qwen2.5-Coder-32B leads
Gemma 2B: 23.0 (#294), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 958 | 1205 |
| LMArena Chinese | 986 | 1222 |
| LMArena Russian | 937 | 1228 |
Instruction Following Qwen2.5-Coder-32B leads
Gemma 2B: 48.5 (#302), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 970 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Gemma 2B: 29.9 (#291), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 981 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Gemma 2B: 24.0 (#308), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Gemma 2B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1002 | 1230 |
| LMArena Creative Writing | 987 | 1174 |
| LMArena Multi-Turn | 945 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Gemma 2B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.6 on the Noometry Index.
Is Gemma 2B or Qwen2.5-Coder-32B better for coding?
Gemma 2B scores higher on coding benchmarks: 29.4 versus 22.6 in the Noometry coding category.
How many benchmarks do Gemma 2B and Qwen2.5-Coder-32B share?
19 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Qwen2.5-Coder-32B has 31.