Model comparison
Gemma 2B vs Gemma 7B
Gemma 2B and Gemma 7B score almost the same on the Noometry Index (29.6 vs 30.0), so choose on price, context window or the category you care about most.
Last verified . 23 shared benchmarks.
GoogleSummary
- They share 23 benchmarks with published results for both. Gemma 2B scores higher in 0 categories and Gemma 7B in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemma 7B leads 27.1 to 24.0.
Side by side
| Gemma 2B | Gemma 7B | |
|---|---|---|
| Provider | ||
| Noometry Index | 29.6 | 30.0 |
| Released | 2024-02-21 | 2024-02-21 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 23 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemma 7B leads
Gemma 2B: 29.4 (#305), Gemma 7B: 30.5 (#294)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| LMArena Coding | 1010 | 1048 |
| HumanEval+ | 20.7% | 28.7% |
| MBPP+ | 34.1% | 43.4% |
Reasoning Gemma 7B leads
Gemma 2B: 18.8 (#275), Gemma 7B: 19.9 (#249)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| LMArena Hard Prompts | 989 | 1042 |
| BIG-Bench Hard | 35.2% | 55.1% |
| Epoch Capabilities Index | 94.2 | 111.99 |
| HellaSwag | 71.4% | 82.2% |
| PIQA | 77.3% | 81.2% |
| WinoGrande | 65.4% | 79% |
| Adversarial NLI | — | 48.7% |
Math Gemma 7B leads
Gemma 2B: 30.0 (#239), Gemma 7B: 31.2 (#228)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| LMArena Math | 1009 | 1066 |
| GSM8K | 17.7% | 46.4% |
Knowledge Not comparable
Gemma 2B: —, Gemma 7B: 27.3 (#252)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| ARC (AI2) Challenge | 42.1% | 78.3% |
| BoolQ | 69.4% | 83.2% |
| MMLU | 42.3% | 66.1% |
| TriviaQA | 53.2% | 72.3% |
| LMArena Expert | — | 1001 |
| OpenBookQA | — | 78.6% |
Multilingual Gemma 7B leads
Gemma 2B: 23.0 (#294), Gemma 7B: 25.1 (#287)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| LMArena Non-English | 958 | 999 |
| LMArena Chinese | 986 | 1035 |
| LMArena Russian | 937 | 993 |
| LMArena French | — | 1025 |
Instruction Following Gemma 7B leads
Gemma 2B: 48.5 (#302), Gemma 7B: 51.5 (#295)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| LMArena Instruction Following | 970 | 1017 |
Long Context Gemma 7B leads
Gemma 2B: 29.9 (#291), Gemma 7B: 31.1 (#282)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| LMArena Longer Query | 981 | 1022 |
Writing & Preference Gemma 7B leads
Gemma 2B: 24.0 (#308), Gemma 7B: 27.1 (#302)
| Benchmark | Gemma 2B | Gemma 7B |
|---|---|---|
| LMArena Text | 1002 | 1056 |
| LMArena Creative Writing | 987 | 1024 |
| LMArena Multi-Turn | 945 | 963 |
Frequently asked questions
Is Gemma 2B better than Gemma 7B?
Gemma 2B and Gemma 7B score almost the same on the Noometry Index (29.6 vs 30.0), so choose on price, context window or the category you care about most.
Is Gemma 2B or Gemma 7B better for coding?
Gemma 7B scores higher on coding benchmarks: 30.5 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2B and Gemma 7B share?
23 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Gemma 7B has 27.