Model comparison
Gemma 7B vs StarCoder 2 3B
Gemma 7B has enough public results to be ranked (#295); StarCoder 2 3B does not yet, so treat this comparison as directional.
Last verified . 6 shared benchmarks.
Google
NVIDIASummary
- They share 6 benchmarks with published results for both. Gemma 7B scores higher in 0 categories and StarCoder 2 3B in 1 category; one gap is clear of the uncertainty.
Side by side
| Gemma 7B | StarCoder 2 3B | |
|---|---|---|
| Provider | NVIDIA | |
| Noometry Index | 29.5 | 35.8 |
| Released | 2024-02-21 | 2024-02-22 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 27 | 7 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding StarCoder 2 3B leads
Gemma 7B: 29.9 (#290), StarCoder 2 3B: 32.9
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| HumanEval+ | 28.7% | 27.4% |
| LMArena Coding | 1048 | — |
| BigCodeBench Complete | — | 21.4% |
| MBPP+ | 43.4% | — |
Reasoning Not comparable
Gemma 7B: 19.0 (#247), StarCoder 2 3B: —
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| Epoch Capabilities Index | 111.74 | 88.01 |
| WinoGrande | 79% | 57.1% |
| LMArena Hard Prompts | 1042 | — |
| Adversarial NLI | 48.7% | — |
| BIG-Bench Hard | 55.1% | — |
| HellaSwag | 82.2% | — |
| PIQA | 81.2% | — |
Math Not comparable
Gemma 7B: 28.7 (#225), StarCoder 2 3B: —
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| GSM8K | 46.4% | 21.6% |
| LMArena Math | 1066 | — |
Knowledge Not comparable
Gemma 7B: 26.8 (#248), StarCoder 2 3B: —
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| ARC (AI2) Challenge | 78.3% | 34.2% |
| MMLU | 66.1% | 36.6% |
| LMArena Expert | 1002 | — |
| BoolQ | 83.2% | — |
| OpenBookQA | 78.6% | — |
| TriviaQA | 72.3% | — |
Multilingual Not comparable
Gemma 7B: 25.3 (#282), StarCoder 2 3B: —
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| LMArena Non-English | 999 | — |
| LMArena Chinese | 1035 | — |
| LMArena French | 1024 | — |
| LMArena Russian | 993 | — |
Instruction Following Not comparable
Gemma 7B: 51.7 (#290), StarCoder 2 3B: —
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| LMArena Instruction Following | 1017 | — |
Long Context Not comparable
Gemma 7B: 31.1 (#277), StarCoder 2 3B: —
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| LMArena Longer Query | 1022 | — |
Writing & Preference Not comparable
Gemma 7B: 27.2 (#297), StarCoder 2 3B: —
| Benchmark | Gemma 7B | StarCoder 2 3B |
|---|---|---|
| LMArena Text | 1056 | — |
| LMArena Creative Writing | 1024 | — |
| LMArena Multi-Turn | 963 | — |
Frequently asked questions
Is Gemma 7B better than StarCoder 2 3B?
Gemma 7B has enough public results to be ranked (#295); StarCoder 2 3B does not yet, so treat this comparison as directional.
Is Gemma 7B or StarCoder 2 3B better for coding?
StarCoder 2 3B scores higher on coding benchmarks: 32.9 versus 29.9 in the Noometry coding category.
How many benchmarks do Gemma 7B and StarCoder 2 3B share?
6 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and StarCoder 2 3B has 7.