Model comparison
Gemma 7B vs StarCoder 2 15B
Gemma 7B has enough public results to be ranked (#299); StarCoder 2 15B 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 15B in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where StarCoder 2 15B leads 36.6 to 30.5.
Side by side
| Gemma 7B | StarCoder 2 15B | |
|---|---|---|
| Provider | NVIDIA | |
| Noometry Index | 30.0 | 38.1 |
| Released | 2024-02-21 | 2024-02-20 |
| 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 15B leads
Gemma 7B: 30.5 (#294), StarCoder 2 15B: 36.6
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| HumanEval+ | 28.7% | 37.8% |
| LMArena Coding | 1048 | — |
| BigCodeBench Complete | — | 38.4% |
| MBPP+ | 43.4% | — |
Reasoning Not comparable
Gemma 7B: 19.9 (#249), StarCoder 2 15B: —
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| Epoch Capabilities Index | 111.99 | 105 |
| WinoGrande | 79% | 64.3% |
| LMArena Hard Prompts | 1042 | — |
| Adversarial NLI | 48.7% | — |
| BIG-Bench Hard | 55.1% | — |
| HellaSwag | 82.2% | — |
| PIQA | 81.2% | — |
Math Not comparable
Gemma 7B: 31.2 (#228), StarCoder 2 15B: —
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| GSM8K | 46.4% | 57.7% |
| LMArena Math | 1066 | — |
Knowledge Not comparable
Gemma 7B: 27.3 (#252), StarCoder 2 15B: —
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| ARC (AI2) Challenge | 78.3% | 47.2% |
| MMLU | 66.1% | 64.1% |
| LMArena Expert | 1001 | — |
| BoolQ | 83.2% | — |
| OpenBookQA | 78.6% | — |
| TriviaQA | 72.3% | — |
Multilingual Not comparable
Gemma 7B: 25.1 (#287), StarCoder 2 15B: —
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| LMArena Non-English | 999 | — |
| LMArena Chinese | 1035 | — |
| LMArena French | 1025 | — |
| LMArena Russian | 993 | — |
Instruction Following Not comparable
Gemma 7B: 51.5 (#295), StarCoder 2 15B: —
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| LMArena Instruction Following | 1017 | — |
Long Context Not comparable
Gemma 7B: 31.1 (#282), StarCoder 2 15B: —
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| LMArena Longer Query | 1022 | — |
Writing & Preference Not comparable
Gemma 7B: 27.1 (#302), StarCoder 2 15B: —
| Benchmark | Gemma 7B | StarCoder 2 15B |
|---|---|---|
| LMArena Text | 1056 | — |
| LMArena Creative Writing | 1024 | — |
| LMArena Multi-Turn | 963 | — |
Frequently asked questions
Is Gemma 7B better than StarCoder 2 15B?
Gemma 7B has enough public results to be ranked (#299); StarCoder 2 15B does not yet, so treat this comparison as directional.
Is Gemma 7B or StarCoder 2 15B better for coding?
StarCoder 2 15B scores higher on coding benchmarks: 36.6 versus 30.5 in the Noometry coding category.
How many benchmarks do Gemma 7B and StarCoder 2 15B share?
6 benchmarks have published results for both models. Gemma 7B has 27 scored results on Noometry and StarCoder 2 15B has 7.