Model comparison
Gemma 3 27B vs Qwen2-72B
Gemma 3 27B and Qwen2-72B score almost the same on the Noometry Index (30.8 vs 30.0), so choose on price, context window or the category you care about most.
Last verified . 20 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Gemma 3 27B scores higher in 5 categories and Qwen2-72B in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 3 27B leads 52.5 to 40.8.
- The biggest single-benchmark swing is MATH Level 5: 74% for Gemma 3 27B and 39.1% for Qwen2-72B.
Side by side
| Gemma 3 27B | Qwen2-72B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 30.8 | 30.0 |
| Released | 2025-03-11 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.08 | — |
| Output $ / M tokens | $0.16 | — |
| Results tracked | 43 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2-72B leads
Gemma 3 27B: 22.5 (#334), Qwen2-72B: 29.1 (#310)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1322 | 1196 |
| Aider Polyglot | 4.9% | — |
| SciCode | 21.2% | — |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| LiveBench Coding | 39.9% | — |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Gemma 3 27B leads
Gemma 3 27B: 25.1 (#110), Qwen2-72B: 17.0 (#146)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 29.5% | — |
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Gemma 3 27B: 16.7 (#301), Qwen2-72B: 23.2 (#181)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1340 | 1191 |
| Epoch Capabilities Index | 130.04 | 125.28 |
| Kagi LLM Benchmark | 40.4% | — |
| CritPt | 0% | — |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | 43.8% | — |
| DTBench | 52.5% | — |
| LiveBench Data Analysis | 51.5% | — |
| LMCA | 12.3% | — |
| LiveBench | 50% | — |
Math Qwen2-72B leads
Gemma 3 27B: 25.9 (#265), Qwen2-72B: 30.2 (#236)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Math | 1312 | 1235 |
| MATH Level 5 | 74% | 39.1% |
| OTIS Mock AIME 2024-2025 | 22.5% | — |
| LiveBench Math | 55.4% | — |
Knowledge Gemma 3 27B leads
Gemma 3 27B: 25.5 (#261), Qwen2-72B: 21.2 (#275)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 47.7% | 40.8% |
| LMArena Expert | 1304 | 1171 |
| Confabulations | 40.3% | — |
| Vectara Hallucination Rate | 7.4% | — |
| MMLU | — | 82.4% |
Multimodal Not comparable
Gemma 3 27B: 32.6 (#100), Qwen2-72B: —
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Vision | 1164 | — |
| GeoBench | 52% | — |
Multilingual Gemma 3 27B leads
Gemma 3 27B: 46.9 (#155), Qwen2-72B: 35.9 (#244)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1334 | 1176 |
| LMArena Chinese | 1346 | 1240 |
| LMArena French | 1368 | 1170 |
| LMArena German | 1362 | 1151 |
| LMArena Japanese | 1287 | 1111 |
| LMArena Korean | 1308 | 1083 |
| LMArena Russian | 1349 | 1169 |
| LMArena Spanish | 1349 | 1169 |
Instruction Following Gemma 3 27B leads
Gemma 3 27B: 70.6 (#160), Qwen2-72B: 61.7 (#241)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1321 | 1181 |
| LiveBench Instruction Following | 74.9% | — |
Long Context Qwen2-72B leads
Gemma 3 27B: 27.6 (#293), Qwen2-72B: 36.1 (#235)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1333 | 1192 |
| Fiction.LiveBench | 33.3% | — |
Writing & Preference Gemma 3 27B leads
Gemma 3 27B: 52.5 (#168), Qwen2-72B: 40.8 (#241)
| Benchmark | Gemma 3 27B | Qwen2-72B |
|---|---|---|
| LMArena Text | 1358 | 1203 |
| LMArena Creative Writing | 1346 | 1181 |
| LMArena Multi-Turn | 1345 | 1196 |
| Short-Story Creative Writing | 79.9% | — |
| EQ-Bench Creative Writing | 1266 | — |
| LiveBench Language | 34.6% | — |
Frequently asked questions
Is Gemma 3 27B better than Qwen2-72B?
Gemma 3 27B and Qwen2-72B score almost the same on the Noometry Index (30.8 vs 30.0), so choose on price, context window or the category you care about most.
Is Gemma 3 27B or Qwen2-72B better for coding?
Qwen2-72B scores higher on coding benchmarks: 29.1 versus 22.5 in the Noometry coding category.
How many benchmarks do Gemma 3 27B and Qwen2-72B share?
20 benchmarks have published results for both models. Gemma 3 27B has 43 scored results on Noometry and Qwen2-72B has 26.