Model comparison
Gemma 7B vs Qwen1.5-32B
Gemma 7B and Qwen1.5-32B score almost the same on the Noometry Index (30.0 vs 30.5), so choose on price, context window or the category you care about most.
Last verified . 14 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 14 benchmarks with published results for both. Gemma 7B scores higher in 1 category and Qwen1.5-32B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemma 7B leads 27.3 to 13.5.
Side by side
| Gemma 7B | Qwen1.5-32B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 30.0 | 30.5 |
| Released | 2024-02-21 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 27 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-32B leads
Gemma 7B: 30.5 (#294), Qwen1.5-32B: 31.7 (#282)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Coding | 1048 | 1155 |
| BigCodeBench Instruct | — | 32.3% |
| BigCodeBench Complete | — | 42% |
| HumanEval+ | 28.7% | — |
| MBPP+ | 43.4% | — |
Reasoning Qwen1.5-32B leads
Gemma 7B: 19.9 (#249), Qwen1.5-32B: 21.8 (#212)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Hard Prompts | 1042 | 1130 |
| Adversarial NLI | 48.7% | — |
| BIG-Bench Hard | 55.1% | — |
| Epoch Capabilities Index | 111.99 | — |
| HellaSwag | 82.2% | — |
| PIQA | 81.2% | — |
| WinoGrande | 79% | — |
Math Qwen1.5-32B leads
Gemma 7B: 31.2 (#228), Qwen1.5-32B: 33.0 (#207)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Math | 1066 | 1155 |
| GSM8K | 46.4% | — |
Knowledge Gemma 7B leads
Gemma 7B: 27.3 (#252), Qwen1.5-32B: 13.5 (#296)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Expert | 1001 | 1126 |
| MMLU | 66.1% | 74.4% |
| GPQA Diamond | — | 30.7% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 83.2% | — |
| OpenBookQA | 78.6% | — |
| TriviaQA | 72.3% | — |
Multilingual Qwen1.5-32B leads
Gemma 7B: 25.1 (#287), Qwen1.5-32B: 31.4 (#259)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Non-English | 999 | 1106 |
| LMArena Chinese | 1035 | 1177 |
| LMArena French | 1025 | 1101 |
| LMArena Russian | 993 | 1073 |
| LMArena German | — | 1058 |
| LMArena Japanese | — | 1027 |
| LMArena Korean | — | 1008 |
| LMArena Spanish | — | 1089 |
Instruction Following Qwen1.5-32B leads
Gemma 7B: 51.5 (#295), Qwen1.5-32B: 57.7 (#265)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Instruction Following | 1017 | 1116 |
Long Context Qwen1.5-32B leads
Gemma 7B: 31.1 (#282), Qwen1.5-32B: 34.7 (#246)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Longer Query | 1022 | 1146 |
Writing & Preference Qwen1.5-32B leads
Gemma 7B: 27.1 (#302), Qwen1.5-32B: 34.2 (#271)
| Benchmark | Gemma 7B | Qwen1.5-32B |
|---|---|---|
| LMArena Text | 1056 | 1137 |
| LMArena Creative Writing | 1024 | 1083 |
| LMArena Multi-Turn | 963 | 1140 |
Frequently asked questions
Is Gemma 7B better than Qwen1.5-32B?
Gemma 7B and Qwen1.5-32B score almost the same on the Noometry Index (30.0 vs 30.5), so choose on price, context window or the category you care about most.
Is Gemma 7B or Qwen1.5-32B better for coding?
Qwen1.5-32B scores higher on coding benchmarks: 31.7 versus 30.5 in the Noometry coding category.
How many benchmarks do Gemma 7B and Qwen1.5-32B share?
14 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and Qwen1.5-32B has 21.