Model comparison
Gemma 7B vs Qwen1.5-110B
Qwen1.5-110B is the stronger model overall, scoring 33.6 to 29.5 on the Noometry Index.
Last verified . 13 shared benchmarks.
Google
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Gemma 7B scores higher in 0 categories and Qwen1.5-110B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen1.5-110B leads 38.1 to 27.2.
Side by side
| Gemma 7B | Qwen1.5-110B | |
|---|---|---|
| Provider | Alibaba (Qwen) | |
| Noometry Index | 29.5 | 33.6 |
| Released | 2024-02-21 | 2024-04-25 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 27 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-110B leads
Gemma 7B: 29.9 (#290), Qwen1.5-110B: 32.4 (#260)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Coding | 1048 | 1184 |
| BigCodeBench Instruct | — | 35% |
| BigCodeBench Complete | — | 44.4% |
| HumanEval+ | 28.7% | — |
| MBPP+ | 43.4% | — |
Reasoning Qwen1.5-110B leads
Gemma 7B: 19.0 (#247), Qwen1.5-110B: 21.7 (#192)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Hard Prompts | 1042 | 1168 |
| Adversarial NLI | 48.7% | — |
| BIG-Bench Hard | 55.1% | — |
| Epoch Capabilities Index | 111.74 | — |
| ForecastBench | — | 57.7 |
| HellaSwag | 82.2% | — |
| PIQA | 81.2% | — |
| WinoGrande | 79% | — |
Math Qwen1.5-110B leads
Gemma 7B: 28.7 (#225), Qwen1.5-110B: 31.1 (#199)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Math | 1066 | 1185 |
| GSM8K | 46.4% | — |
Knowledge Qwen1.5-110B leads
Gemma 7B: 26.8 (#248), Qwen1.5-110B: 30.7 (#217)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Expert | 1002 | 1144 |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 83.2% | — |
| MMLU | 66.1% | — |
| OpenBookQA | 78.6% | — |
| TriviaQA | 72.3% | — |
Multilingual Qwen1.5-110B leads
Gemma 7B: 25.3 (#282), Qwen1.5-110B: 33.8 (#245)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Non-English | 999 | 1142 |
| LMArena Chinese | 1035 | 1206 |
| LMArena French | 1024 | 1150 |
| LMArena Russian | 993 | 1118 |
| LMArena German | — | 1123 |
| LMArena Japanese | — | 1074 |
| LMArena Korean | — | 1044 |
| LMArena Spanish | — | 1141 |
Instruction Following Qwen1.5-110B leads
Gemma 7B: 51.7 (#290), Qwen1.5-110B: 60.6 (#247)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Instruction Following | 1017 | 1158 |
Long Context Qwen1.5-110B leads
Gemma 7B: 31.1 (#277), Qwen1.5-110B: 35.0 (#237)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Longer Query | 1022 | 1157 |
Writing & Preference Qwen1.5-110B leads
Gemma 7B: 27.2 (#297), Qwen1.5-110B: 38.1 (#250)
| Benchmark | Gemma 7B | Qwen1.5-110B |
|---|---|---|
| LMArena Text | 1056 | 1175 |
| LMArena Creative Writing | 1024 | 1148 |
| LMArena Multi-Turn | 963 | 1160 |
Frequently asked questions
Is Gemma 7B better than Qwen1.5-110B?
Qwen1.5-110B is the stronger model overall, scoring 33.6 to 29.5 on the Noometry Index.
Is Gemma 7B or Qwen1.5-110B better for coding?
Qwen1.5-110B scores higher on coding benchmarks: 32.4 versus 29.9 in the Noometry coding category.
How many benchmarks do Gemma 7B and Qwen1.5-110B share?
13 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and Qwen1.5-110B has 20.