Model comparison
Gemma 2 27B vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 29.4 on the Noometry Index. Gemma 2 27B costs 3.8× less per token, which makes it the better buy when Qwen2.5 72B Instruct's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. Gemma 2 27B scores higher in 1 category and Qwen2.5 72B Instruct in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5 72B Instruct leads 19.3 to 10.7.
- The biggest single-benchmark swing is MATH Level 5: 27.9% for Gemma 2 27B and 63.2% for Qwen2.5 72B Instruct.
- Gemma 2 27B is cheaper at $0.65 / $0.65 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct accepts more context: 131K tokens versus 8K.
Side by side
| Gemma 2 27B | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 29.4 | 31.9 |
| Released | 2024-06-24 | 2024-09 |
| Weights | Open | Open |
| Context window | 8K | 131K |
| Max output | 2K | 8K |
| Input $ / M tokens | $0.65 | $1.40 |
| Output $ / M tokens | $0.65 | $5.60 |
| Results tracked | 34 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemma 2 27B: 34.1 (#246), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| BigCodeBench Instruct | 42.8% | 45.8% |
| LMArena Coding | 1211 | 1292 |
| BigCodeBench Complete | 52.5% | 55.9% |
| WeirdML | — | 16% |
| LiveBench Coding | 36% | — |
Agentic & Tool Use Not comparable
Gemma 2 27B: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Gemma 2 27B: 15.3 (#315), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1198 | 1271 |
| DTBench | 48% | 62.9% |
| LMCA | 7.1% | 13.4% |
| Epoch Capabilities Index | 122.08 | 129 |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 47.9% | — |
| BIG-Bench Hard | — | 79.8% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| LiveBench | 38.2% | — |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |
Math Qwen2.5 72B Instruct leads
Gemma 2 27B: 10.7 (#311), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.4% | 8.1% |
| LMArena Math | 1212 | 1283 |
| MATH Level 5 | 27.9% | 63.2% |
| Omni-MATH | — | 33% |
| LiveBench Math | 26.5% | — |
Knowledge Qwen2.5 72B Instruct leads
Gemma 2 27B: 19.0 (#280), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 36.5% | 49.1% |
| Confabulations | 27.1% | 19.1% |
| LMArena Expert | 1172 | 1245 |
| MMLU | 75.7% | 85.3% |
| MMLU-Pro | — | 63.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| TriviaQA | — | 71.9% |
Multilingual Qwen2.5 72B Instruct leads
Gemma 2 27B: 38.6 (#226), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1217 | 1252 |
| LMArena Chinese | 1221 | 1272 |
| LMArena French | 1247 | 1280 |
| LMArena German | 1209 | 1234 |
| LMArena Japanese | 1175 | 1180 |
| LMArena Korean | 1174 | 1188 |
| LMArena Russian | 1234 | 1264 |
| LMArena Spanish | 1228 | 1256 |
Instruction Following Qwen2.5 72B Instruct leads
Gemma 2 27B: 60.5 (#249), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1206 | 1254 |
| LiveBench Instruction Following | 58.1% | — |
| IFEval | — | 80.6% |
Long Context Qwen2.5 72B Instruct leads
Gemma 2 27B: 37.3 (#218), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1231 | 1282 |
Writing & Preference Qwen2.5 72B Instruct leads
Gemma 2 27B: 44.2 (#225), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Gemma 2 27B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1231 | 1269 |
| LMArena Creative Writing | 1241 | 1221 |
| LMArena Multi-Turn | 1224 | 1272 |
| WildBench | — | 80.2% |
| LiveBench Language | 32.6% | — |
Frequently asked questions
Is Gemma 2 27B better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 29.4 on the Noometry Index. Gemma 2 27B costs 3.8× less per token, which makes it the better buy when Qwen2.5 72B Instruct's lead doesn't matter for your workload.
Which is cheaper, Gemma 2 27B or Qwen2.5 72B Instruct?
Gemma 2 27B is cheaper. It lists at $0.65 per million input tokens and $0.65 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.
Is Gemma 2 27B or Qwen2.5 72B Instruct better for coding?
They score almost the same on coding (34.1 vs 33.2); test both on your own repository before choosing.
Which has the bigger context window?
Qwen2.5 72B Instruct does, with 131K tokens against 8K.
How many benchmarks do Gemma 2 27B and Qwen2.5 72B Instruct share?
27 benchmarks have published results for both models. Gemma 2 27B has 34 scored results on Noometry and Qwen2.5 72B Instruct has 43.