Model comparison
Gemma 2B vs Phi-3.5-mini
Gemma 2B has enough public results to be ranked (#307); Phi-3.5-mini does not yet, so treat this comparison as directional.
Last verified . 3 shared benchmarks.
Google
MicrosoftSummary
- They share 3 benchmarks with published results for both. Gemma 2B scores higher in 0 categories and Phi-3.5-mini in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where Phi-3.5-mini leads 34.6 to 29.4.
Side by side
| Gemma 2B | Phi-3.5-mini | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 29.6 | 36.9 |
| Released | 2024-02-21 | 2024-08-16 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 23 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-3.5-mini leads
Gemma 2B: 29.4 (#305), Phi-3.5-mini: 34.6 (#232)
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| BigCodeBench Instruct | — | 32.8% |
| LMArena Coding | 1010 | — |
| BigCodeBench Complete | — | 38.5% |
| HumanEval+ | 20.7% | — |
| MBPP+ | 34.1% | — |
Reasoning Not comparable
Gemma 2B: 18.8 (#275), Phi-3.5-mini: —
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| PIQA | 77.3% | 81% |
| LMArena Hard Prompts | 989 | — |
| BIG-Bench Hard | 35.2% | — |
| Epoch Capabilities Index | 94.2 | — |
| HellaSwag | 71.4% | — |
| WinoGrande | 65.4% | — |
Math Not comparable
Gemma 2B: 30.0 (#239), Phi-3.5-mini: —
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| GSM8K | 17.7% | 86.2% |
| LMArena Math | 1009 | — |
Knowledge Not comparable
Gemma 2B: —, Phi-3.5-mini: —
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| BoolQ | 69.4% | 78% |
| ARC (AI2) Challenge | 42.1% | — |
| MMLU | 42.3% | — |
| TriviaQA | 53.2% | — |
Multilingual Not comparable
Gemma 2B: 23.0 (#294), Phi-3.5-mini: —
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| LMArena Non-English | 958 | — |
| LMArena Chinese | 986 | — |
| LMArena Russian | 937 | — |
Instruction Following Not comparable
Gemma 2B: 48.5 (#302), Phi-3.5-mini: —
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| LMArena Instruction Following | 970 | — |
Long Context Not comparable
Gemma 2B: 29.9 (#291), Phi-3.5-mini: —
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| LMArena Longer Query | 981 | — |
Writing & Preference Not comparable
Gemma 2B: 24.0 (#308), Phi-3.5-mini: —
| Benchmark | Gemma 2B | Phi-3.5-mini |
|---|---|---|
| LMArena Text | 1002 | — |
| LMArena Creative Writing | 987 | — |
| LMArena Multi-Turn | 945 | — |
Frequently asked questions
Is Gemma 2B better than Phi-3.5-mini?
Gemma 2B has enough public results to be ranked (#307); Phi-3.5-mini does not yet, so treat this comparison as directional.
Is Gemma 2B or Phi-3.5-mini better for coding?
Phi-3.5-mini scores higher on coding benchmarks: 34.6 versus 29.4 in the Noometry coding category.
How many benchmarks do Gemma 2B and Phi-3.5-mini share?
3 benchmarks have published results for both models. Gemma 2B has 23 scored results on Noometry and Phi-3.5-mini has 5.