Model comparison
Gemini 2.0 Flash (Feb 2025) vs phi-3-medium 14B
Gemini 2.0 Flash (Feb 2025) is the stronger model overall, scoring 34.8 to 29.1 on the Noometry Index.
Last verified . 6 shared benchmarks.
Google
MicrosoftSummary
- They share 6 benchmarks with published results for both. Gemini 2.0 Flash (Feb 2025) scores higher in 2 categories and phi-3-medium 14B in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 2.0 Flash (Feb 2025) leads 32.0 to 9.1.
- The biggest single-benchmark swing is MATH Level 5: 82.2% for Gemini 2.0 Flash (Feb 2025) and 17.6% for phi-3-medium 14B.
- phi-3-medium 14B has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 34.8 | 29.1 |
| Released | 2024-12-06 | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 54 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding phi-3-medium 14B leads
Gemini 2.0 Flash (Feb 2025): 28.2 (#308), phi-3-medium 14B: 36.3 (#198)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| BigCodeBench Instruct | 45.9% | 37.6% |
| BigCodeBench Complete | 59.9% | 48.7% |
| SWE-bench Verified (bash only) | 13.5% | — |
| Aider Polyglot | 38.2% | — |
| WeirdML | 25.8% | — |
| LiveBench Coding | 63.4% | — |
| LMArena Coding | 1351 | — |
| CadEval | 30% | — |
Agentic & Tool Use Not comparable
Gemini 2.0 Flash (Feb 2025): 28.0 (#85), phi-3-medium 14B: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| TheAgentCompany | 11.4% | — |
Reasoning Not comparable
Gemini 2.0 Flash (Feb 2025): 15.2 (#308), phi-3-medium 14B: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| Epoch Capabilities Index | 135.36 | 121.18 |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 31.1% | — |
| Kagi LLM Benchmark | 37.8% | — |
| EnigmaEval | 1.1% | — |
| LiveBench Reasoning | 78.2% | — |
| LMArena Hard Prompts | 1346 | — |
| DTBench | 63.2% | — |
| LiveBench Data Analysis | 69.4% | — |
| Adversarial NLI | — | 55.8% |
| BIG-Bench Hard | — | 81.4% |
| HellaSwag | — | 82.4% |
| LiveBench | 66.9% | — |
| WinoGrande | — | 81.5% |
Math Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 36.1 (#127), phi-3-medium 14B: 25.2 (#249)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| MATH Level 5 | 82.2% | 17.6% |
| OTIS Mock AIME 2024-2025 | 57.8% | — |
| Omni-MATH | 45.9% | — |
| LiveBench Math | 75.8% | — |
| LMArena Math | 1351 | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 32.0 (#207), phi-3-medium 14B: 9.1 (#301)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| GPQA Diamond | 64.1% | 27.6% |
| MMLU | 79.7% | 78% |
| Humanity's Last Exam | 6.6% | — |
| MMLU-Pro | 73.7% | — |
| Confabulations | 12.4% | — |
| GPQA (HELM) | 55.6% | — |
| LMArena Expert | 1340 | — |
| ARC (AI2) Challenge | — | 91.6% |
| OpenBookQA | — | 87.4% |
| TriviaQA | — | 73.9% |
Multimodal Not comparable
Gemini 2.0 Flash (Feb 2025): 35.9 (#74), phi-3-medium 14B: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| LMArena Vision | 1158 | — |
| GeoBench | 77% | — |
Multilingual Not comparable
Gemini 2.0 Flash (Feb 2025): 47.6 (#144), phi-3-medium 14B: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| LMArena Non-English | 1342 | — |
| LMArena Chinese | 1373 | — |
| LMArena French | 1390 | — |
| LMArena German | 1354 | — |
| LMArena Japanese | 1293 | — |
| LMArena Korean | 1314 | — |
| LMArena Russian | 1351 | — |
| LMArena Spanish | 1363 | — |
Instruction Following Not comparable
Gemini 2.0 Flash (Feb 2025): 74.6 (#91), phi-3-medium 14B: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| LiveBench Instruction Following | 85.8% | — |
| IFEval | 84.1% | — |
| LMArena Instruction Following | 1336 | — |
Long Context Not comparable
Gemini 2.0 Flash (Feb 2025): 38.1 (#198), phi-3-medium 14B: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| Fiction.LiveBench | 61.1% | — |
| LMArena Longer Query | 1345 | — |
Writing & Preference Not comparable
Gemini 2.0 Flash (Feb 2025): 49.8 (#184), phi-3-medium 14B: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | phi-3-medium 14B |
|---|---|---|
| LMArena Text | 1354 | — |
| LMArena Creative Writing | 1340 | — |
| Short-Story Creative Writing | 73.8% | — |
| EQ-Bench Creative Writing | 1128 | — |
| WildBench | 80% | — |
| LMArena Multi-Turn | 1350 | — |
| LiveBench Language | 51.3% | — |
Frequently asked questions
Is Gemini 2.0 Flash (Feb 2025) better than phi-3-medium 14B?
Gemini 2.0 Flash (Feb 2025) is the stronger model overall, scoring 34.8 to 29.1 on the Noometry Index.
Is Gemini 2.0 Flash (Feb 2025) or phi-3-medium 14B better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.3 versus 28.2 in the Noometry coding category.
How many benchmarks do Gemini 2.0 Flash (Feb 2025) and phi-3-medium 14B share?
6 benchmarks have published results for both models. Gemini 2.0 Flash (Feb 2025) has 54 scored results on Noometry and phi-3-medium 14B has 13.