Model comparison
Gemini 2.0 Flash (Feb 2025) vs Phi-4
Gemini 2.0 Flash (Feb 2025) is the stronger model overall, scoring 35.1 to 31.2 on the Noometry Index.
Last verified . 33 shared benchmarks.
Google
MicrosoftSummary
- They share 33 benchmarks with published results for both. Gemini 2.0 Flash (Feb 2025) scores higher in 6 categories and Phi-4 in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.0 Flash (Feb 2025) leads 37.9 to 20.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 57.8% for Gemini 2.0 Flash (Feb 2025) and 13.8% for Phi-4.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.0 Flash (Feb 2025) | Phi-4 | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 35.1 | 31.2 |
| Released | 2024-12-06 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 54 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-4 leads
Gemini 2.0 Flash (Feb 2025): 28.4 (#315), Phi-4: 34.4 (#239)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 45.9% | 45.5% |
| LiveBench Coding | 63.4% | 30.7% |
| LMArena Coding | 1350 | 1231 |
| BigCodeBench Complete | 59.9% | 55.4% |
| SWE-bench Verified (bash only) | 13.5% | — |
| Aider Polyglot | 38.2% | — |
| WeirdML | 25.8% | — |
| CadEval | 30% | — |
Agentic & Tool Use Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 28.1 (#92), Phi-4: 22.8 (#128)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| TheAgentCompany | 11.4% | — |
| BALROG | — | 11.6% |
Reasoning Phi-4 leads
Gemini 2.0 Flash (Feb 2025): 15.2 (#318), Phi-4: 17.7 (#291)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| LiveBench Reasoning | 78.2% | 47.8% |
| LMArena Hard Prompts | 1346 | 1220 |
| LiveBench Data Analysis | 69.4% | 45.2% |
| Epoch Capabilities Index | 135.36 | 130.42 |
| LiveBench | 66.9% | 41.6% |
| ARC-AGI-2 | 1.3% | — |
| SimpleBench | 31.1% | — |
| Kagi LLM Benchmark | 37.8% | — |
| Chess Puzzles | — | 1% |
| EnigmaEval | 1.1% | — |
| DTBench | 63.2% | — |
Math Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 37.9 (#146), Phi-4: 20.8 (#285)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 13.8% |
| LiveBench Math | 75.8% | 42% |
| LMArena Math | 1352 | 1246 |
| MATH Level 5 | 82.2% | 64.9% |
| Omni-MATH | 45.9% | — |
| FrontierMath (Feb 2025 set) | 1.7% | — |
Knowledge Too close to call
Gemini 2.0 Flash (Feb 2025): 32.0 (#213), Phi-4: 32.6 (#209)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| GPQA Diamond | 64.1% | 56.1% |
| Confabulations | 12.4% | 29.4% |
| LMArena Expert | 1339 | 1203 |
| MMLU | 79.7% | 84.8% |
| Humanity's Last Exam | 6.6% | — |
| MMLU-Pro | 73.7% | — |
| Vectara Hallucination Rate | — | 3.7% |
| GPQA (HELM) | 55.6% | — |
Multimodal Not comparable
Gemini 2.0 Flash (Feb 2025): 36.5 (#79), Phi-4: —
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| LMArena Vision | 1158 | — |
| GeoBench | 77% | — |
Multilingual Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 47.4 (#149), Phi-4: 37.2 (#237)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| LMArena Non-English | 1342 | 1197 |
| LMArena Chinese | 1373 | 1212 |
| LMArena French | 1391 | 1224 |
| LMArena German | 1353 | 1222 |
| LMArena Japanese | 1294 | 1158 |
| LMArena Korean | 1313 | 1151 |
| LMArena Russian | 1351 | 1209 |
| LMArena Spanish | 1363 | 1234 |
Instruction Following Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 74.4 (#97), Phi-4: 60.4 (#251)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| LiveBench Instruction Following | 85.8% | 58.4% |
| LMArena Instruction Following | 1336 | 1201 |
| IFEval | 84.1% | — |
Long Context Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 38.1 (#203), Phi-4: 36.9 (#226)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1344 | 1217 |
| Fiction.LiveBench | 61.1% | — |
Writing & Preference Gemini 2.0 Flash (Feb 2025) leads
Gemini 2.0 Flash (Feb 2025): 49.5 (#190), Phi-4: 40.5 (#244)
| Benchmark | Gemini 2.0 Flash (Feb 2025) | Phi-4 |
|---|---|---|
| LMArena Text | 1354 | 1217 |
| LMArena Creative Writing | 1340 | 1182 |
| Short-Story Creative Writing | 73.8% | 62.6% |
| LMArena Multi-Turn | 1350 | 1206 |
| LiveBench Language | 51.3% | 25.6% |
| EQ-Bench Creative Writing | 1128 | — |
| WildBench | 80% | — |
Frequently asked questions
Is Gemini 2.0 Flash (Feb 2025) better than Phi-4?
Gemini 2.0 Flash (Feb 2025) is the stronger model overall, scoring 35.1 to 31.2 on the Noometry Index.
Is Gemini 2.0 Flash (Feb 2025) or Phi-4 better for coding?
Phi-4 scores higher on coding benchmarks: 34.4 versus 28.4 in the Noometry coding category.
How many benchmarks do Gemini 2.0 Flash (Feb 2025) and Phi-4 share?
33 benchmarks have published results for both models. Gemini 2.0 Flash (Feb 2025) has 54 scored results on Noometry and Phi-4 has 37.