Model comparison
Gemma 2 27B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.4 on the Noometry Index.
Last verified . 23 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 23 benchmarks with published results for both. Gemma 2 27B scores higher in 3 categories and Qwen2.5-Coder-32B in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Coder-32B leads 33.3 to 10.7.
- The biggest single-benchmark swing is LiveBench Coding: 36% for Gemma 2 27B and 56.9% for Qwen2.5-Coder-32B.
- Gemma 2 27B is cheaper at $0.65 / $0.65 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Qwen2.5-Coder-32B accepts more context: 33K tokens versus 8K.
Side by side
| Gemma 2 27B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 29.4 | 33.4 |
| Released | 2024-06-24 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 8K | 33K |
| Max output | 2K | 29K |
| Input $ / M tokens | $0.65 | $0.66 |
| Output $ / M tokens | $0.65 | $1 |
| Results tracked | 34 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 2 27B leads
Gemma 2 27B: 34.1 (#246), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| BigCodeBench Instruct | 42.8% | 49% |
| LiveBench Coding | 36% | 56.9% |
| LMArena Coding | 1211 | 1276 |
| BigCodeBench Complete | 52.5% | 58% |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Qwen2.5-Coder-32B leads
Gemma 2 27B: 15.3 (#315), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Reasoning | 28.1% | 42.1% |
| LMArena Hard Prompts | 1198 | 1251 |
| LiveBench Data Analysis | 47.9% | 49.9% |
| Epoch Capabilities Index | 122.08 | 119.49 |
| LiveBench | 38.2% | 46.2% |
| DTBench | 48% | — |
| LMCA | 7.1% | — |
| HellaSwag | — | 83% |
| WinoGrande | — | 80.8% |
Math Qwen2.5-Coder-32B leads
Gemma 2 27B: 10.7 (#311), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Math | 26.5% | 46.6% |
| LMArena Math | 1212 | 1251 |
| OTIS Mock AIME 2024-2025 | 1.4% | — |
| MATH Level 5 | 27.9% | — |
| GSM8K | — | 93% |
Knowledge Qwen2.5-Coder-32B leads
Gemma 2 27B: 19.0 (#280), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1172 | 1221 |
| MMLU | 75.7% | 79.1% |
| GPQA Diamond | 36.5% | — |
| Confabulations | 27.1% | — |
| ARC (AI2) Challenge | — | 70.5% |
Multilingual Too close to call
Gemma 2 27B: 38.6 (#226), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1217 | 1205 |
| LMArena Chinese | 1221 | 1222 |
| LMArena Russian | 1234 | 1228 |
| LMArena French | 1247 | — |
| LMArena German | 1209 | — |
| LMArena Japanese | 1175 | — |
| LMArena Korean | 1174 | — |
| LMArena Spanish | 1228 | — |
Instruction Following Too close to call
Gemma 2 27B: 60.5 (#249), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| LiveBench Instruction Following | 58.1% | 58.7% |
| LMArena Instruction Following | 1206 | 1223 |
Long Context Too close to call
Gemma 2 27B: 37.3 (#218), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1231 | 1251 |
Writing & Preference Gemma 2 27B leads
Gemma 2 27B: 44.2 (#225), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Gemma 2 27B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1231 | 1230 |
| LMArena Creative Writing | 1241 | 1174 |
| LMArena Multi-Turn | 1224 | 1222 |
| LiveBench Language | 32.6% | 23.3% |
Frequently asked questions
Is Gemma 2 27B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.4 on the Noometry Index.
Which is cheaper, Gemma 2 27B or Qwen2.5-Coder-32B?
Gemma 2 27B is cheaper. It lists at $0.65 per million input tokens and $0.65 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Gemma 2 27B or Qwen2.5-Coder-32B better for coding?
Gemma 2 27B scores higher on coding benchmarks: 34.1 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Qwen2.5-Coder-32B does, with 33K tokens against 8K.
How many benchmarks do Gemma 2 27B and Qwen2.5-Coder-32B share?
23 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Qwen2.5-Coder-32B has 31.