Model comparison
Gemini 1.5 Pro (May 2024) vs Phi-2
Gemini 1.5 Pro (May 2024) has enough public results to be ranked (#261); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 5 shared benchmarks.
Google
MicrosoftSummary
- They share 5 benchmarks with published results for both.
- Phi-2 has downloadable open weights; the other is API-only.
Side by side
| Gemini 1.5 Pro (May 2024) | Phi-2 | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 32.1 | — |
| Released | 2024-02-15 | 2023-12-12 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 45 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Gemini 1.5 Pro (May 2024): 34.2 (#241), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| HumanEval+ | 79.3% | 45.1% |
| MBPP+ | 74.6% | 54.2% |
| WeirdML | 22.2% | — |
| BigCodeBench Instruct | 43.8% | — |
| LMArena Coding | 1294 | — |
| BigCodeBench Complete | 57.5% | — |
| CadEval | 34% | — |
Agentic & Tool Use Not comparable
Gemini 1.5 Pro (May 2024): 17.9 (#145), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| TheAgentCompany | 3.4% | — |
| Cybench | 7.5% | — |
| BALROG | 21% | — |
Reasoning Not comparable
Gemini 1.5 Pro (May 2024): 12.3 (#338), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| BIG-Bench Hard | 89.2% | 59.4% |
| Epoch Capabilities Index | 131.73 | 107.94 |
| ARC-AGI-2 | 0.8% | — |
| SimpleBench | 27.1% | — |
| LMArena Hard Prompts | 1296 | — |
| DTBench | 59% | — |
| Adversarial NLI | — | 42.5% |
| ForecastBench | 58.4 | — |
| HellaSwag | — | 53.6% |
| WinoGrande | — | 54.7% |
Math Not comparable
Gemini 1.5 Pro (May 2024): 25.8 (#266), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 23.1% | — |
| Omni-MATH | 36.4% | — |
| LMArena Math | 1315 | — |
| MATH Level 5 | 70.4% | — |
Knowledge Not comparable
Gemini 1.5 Pro (May 2024): 29.4 (#239), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| MMLU | 86.9% | 58.4% |
| GPQA Diamond | 57.2% | — |
| Humanity's Last Exam | 4.6% | — |
| MMLU-Pro | 73.7% | — |
| Confabulations | 13.5% | — |
| GPQA (HELM) | 53.4% | — |
| LMArena Expert | 1279 | — |
| ARC (AI2) Challenge | — | 75.9% |
| OpenBookQA | — | 73.6% |
| TriviaQA | — | 45.2% |
Multimodal Not comparable
Gemini 1.5 Pro (May 2024): 36.8 (#77), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| LMArena Vision | 1161 | — |
| Video-MME | 75% | — |
Multilingual Not comparable
Gemini 1.5 Pro (May 2024): 45.3 (#174), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| LMArena Non-English | 1312 | — |
| LMArena Chinese | 1331 | — |
| LMArena French | 1302 | — |
| LMArena German | 1286 | — |
| LMArena Japanese | 1292 | — |
| LMArena Korean | 1298 | — |
| LMArena Russian | 1320 | — |
| LMArena Spanish | 1311 | — |
Instruction Following Not comparable
Gemini 1.5 Pro (May 2024): 68.6 (#185), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| IFEval | 83.7% | — |
| LMArena Instruction Following | 1297 | — |
Long Context Not comparable
Gemini 1.5 Pro (May 2024): 39.8 (#169), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| LMArena Longer Query | 1308 | — |
Writing & Preference Not comparable
Gemini 1.5 Pro (May 2024): 52.4 (#172), Phi-2: —
| Benchmark | Gemini 1.5 Pro (May 2024) | Phi-2 |
|---|---|---|
| LMArena Text | 1319 | — |
| LMArena Creative Writing | 1333 | — |
| WildBench | 81.3% | — |
| LMArena Multi-Turn | 1296 | — |
Frequently asked questions
Is Gemini 1.5 Pro (May 2024) better than Phi-2?
Gemini 1.5 Pro (May 2024) has enough public results to be ranked (#261); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Gemini 1.5 Pro (May 2024) and Phi-2 share?
5 benchmarks have published results for both models. Gemini 1.5 Pro (May 2024) has 45 scored results on Noometry and Phi-2 has 11.