Model comparison
Gemini 3 Pro vs Phi-4
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 31.2 on the Noometry Index.
Last verified . 24 shared benchmarks.
Google
MicrosoftSummary
- They share 24 benchmarks with published results for both. Gemini 3 Pro scores higher in 9 categories and Phi-4 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 17.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 91.4% for Gemini 3 Pro and 13.8% for Phi-4.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Gemini 3 Pro | Phi-4 | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 54.8 | 31.2 |
| Released | 2025-11-18 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 67 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 3 Pro leads
Gemini 3 Pro: 51.6 (#39), Phi-4: 34.4 (#239)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| LMArena Coding | 1481 | 1231 |
| SWE-bench Verified | 72.9% | — |
| SWE-bench Verified (bash only) | 74.2% | — |
| LMArena WebDev | 1440 | — |
| SWE-bench Multilingual | 68.7% | — |
| GSO | 18.6% | — |
| WeirdML | 69.9% | — |
| BigCodeBench Instruct | — | 45.5% |
| LiveBench Coding | — | 30.7% |
| BigCodeBench Complete | — | 55.4% |
| ALE-Bench | 1,177 | — |
| AlgoTune | 1.83 | — |
Agentic & Tool Use Gemini 3 Pro leads
Gemini 3 Pro: 40.6 (#23), Phi-4: 22.8 (#128)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 72.5% | 28.8% |
| BALROG | 58.1% | 11.6% |
| Terminal-Bench | 69.4% | — |
| GDPval | 40.3% | — |
| Remote Labor Index | 1.3% | — |
| τ²-bench Airline | 80.5% | — |
| τ²-bench Banking | 18% | — |
| τ²-bench Retail | 75.9% | — |
| τ²-bench Telecom | 91% | — |
| DeepResearch Bench | 46.3% | — |
| LMArena Search | 1207 | — |
| METR Time Horizons | 71% | — |
| Vending-Bench 2 | 5,478 | — |
Reasoning Gemini 3 Pro leads
Gemini 3 Pro: 52.5 (#31), Phi-4: 17.7 (#291)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| Chess Puzzles | 31% | 1% |
| LMArena Hard Prompts | 1480 | 1220 |
| Epoch Capabilities Index | 152.92 | 130.42 |
| ARC-AGI-2 | 31.1% | — |
| SimpleBench | 76.4% | — |
| Kagi LLM Benchmark | 80.1% | — |
| NYT Connections (extended) | 94.4% | — |
| ARC-AGI-1 | 75% | — |
| CritPt | 6.9% | — |
| EnigmaEval | 18.2% | — |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| ForecastBench | 61.2 | — |
| LiveBench | — | 41.6% |
Math Gemini 3 Pro leads
Gemini 3 Pro: 49.9 (#59), Phi-4: 20.8 (#285)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.4% | 13.8% |
| LMArena Math | 1476 | 1246 |
| MathArena Final-Answer Competitions | 67% | — |
| ProofBench | 20% | — |
| Omni-MATH | 55.5% | — |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
| FrontierMath (Feb 2025 set) | 37.6% | — |
| FrontierMath Tier 4 (v1) | 18.8% | — |
Knowledge Gemini 3 Pro leads
Gemini 3 Pro: 64.4 (#16), Phi-4: 32.6 (#209)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| GPQA Diamond | 92.6% | 56.1% |
| Vectara Hallucination Rate | 13.6% | 3.7% |
| LMArena Expert | 1475 | 1203 |
| Humanity's Last Exam | 37.5% | — |
| MMLU-Pro | 90.3% | — |
| Confabulations | — | 29.4% |
| GPQA (HELM) | 80.3% | — |
| MMLU | — | 84.8% |
Multimodal Not comparable
Gemini 3 Pro: 57.6 (#2), Phi-4: —
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| LMArena Vision | 1305 | — |
| GeoBench | 84% | — |
| VPCT | 91% | — |
| LMArena Document | 1434 | — |
Multilingual Gemini 3 Pro leads
Gemini 3 Pro: 56.9 (#16), Phi-4: 37.2 (#237)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| LMArena Non-English | 1474 | 1197 |
| LMArena Chinese | 1523 | 1212 |
| LMArena French | 1492 | 1224 |
| LMArena German | 1515 | 1222 |
| LMArena Japanese | 1510 | 1158 |
| LMArena Korean | 1448 | 1151 |
| LMArena Russian | 1493 | 1209 |
| LMArena Spanish | 1470 | 1234 |
Instruction Following Gemini 3 Pro leads
Gemini 3 Pro: 76.3 (#45), Phi-4: 60.4 (#251)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1458 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
| IFEval | 87.7% | — |
Long Context Gemini 3 Pro leads
Gemini 3 Pro: 44.0 (#79), Phi-4: 36.9 (#226)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1471 | 1217 |
| CL-bench | 15.8% | — |
Writing & Preference Gemini 3 Pro leads
Gemini 3 Pro: 66.4 (#35), Phi-4: 40.5 (#244)
| Benchmark | Gemini 3 Pro | Phi-4 |
|---|---|---|
| LMArena Text | 1479 | 1217 |
| LMArena Creative Writing | 1482 | 1182 |
| LMArena Multi-Turn | 1484 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| EQ-Bench Creative Writing | 1525 | — |
| WildBench | 85.9% | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Gemini 3 Pro better than Phi-4?
Gemini 3 Pro is the stronger model overall, scoring 54.8 to 31.2 on the Noometry Index.
Is Gemini 3 Pro or Phi-4 better for coding?
Gemini 3 Pro scores higher on coding benchmarks: 51.6 versus 34.4 in the Noometry coding category.
How many benchmarks do Gemini 3 Pro and Phi-4 share?
24 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and Phi-4 has 37.