Model comparison
Gemini 1.5 Flash (May 2024) vs Phi-4
Gemini 1.5 Flash (May 2024) is the stronger model overall, scoring 33.2 to 31.2 on the Noometry Index.
Last verified . 25 shared benchmarks.
Google
MicrosoftSummary
- They share 25 benchmarks with published results for both. Gemini 1.5 Flash (May 2024) scores higher in 8 categories and Phi-4 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 1.5 Flash (May 2024) leads 48.7 to 40.5.
- The biggest single-benchmark swing is GPQA Diamond: 47.3% for Gemini 1.5 Flash (May 2024) and 56.1% for Phi-4.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Gemini 1.5 Flash (May 2024) | Phi-4 | |
|---|---|---|
| Provider | Microsoft | |
| Noometry Index | 33.2 | 31.2 |
| Released | 2024-05-14 | 2024-12-11 |
| Weights | Proprietary | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 42 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 1.5 Flash (May 2024): 34.4 (#236), Phi-4: 34.4 (#239)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 43.5% | 45.5% |
| LMArena Coding | 1261 | 1231 |
| BigCodeBench Complete | 55.1% | 55.4% |
| WeirdML | 24.9% | — |
| LiveBench Coding | — | 30.7% |
| HumanEval+ | 75.6% | — |
| MBPP+ | 67.5% | — |
Agentic & Tool Use Gemini 1.5 Flash (May 2024) leads
Gemini 1.5 Flash (May 2024): 26.6 (#102), Phi-4: 22.8 (#128)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| BALROG | 14.6% | 11.6% |
| Berkeley Function Calling Leaderboard | — | 28.8% |
Reasoning Gemini 1.5 Flash (May 2024) leads
Gemini 1.5 Flash (May 2024): 21.7 (#215), Phi-4: 17.7 (#291)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1220 |
| Epoch Capabilities Index | 129.36 | 130.42 |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| DTBench | 53.8% | — |
| LiveBench Data Analysis | — | 45.2% |
| ForecastBench | 53.9 | — |
| LiveBench | — | 41.6% |
| PIQA | 87.5% | — |
Math Gemini 1.5 Flash (May 2024) leads
Gemini 1.5 Flash (May 2024): 22.1 (#281), Phi-4: 20.8 (#285)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 16.3% | 13.8% |
| LMArena Math | 1269 | 1246 |
| MATH Level 5 | 61.9% | 64.9% |
| Omni-MATH | 30.4% | — |
| LiveBench Math | — | 42% |
| FrontierMath (Feb 2025 set) | 0% | — |
| GSM8K | 82.4% | — |
Knowledge Phi-4 leads
Gemini 1.5 Flash (May 2024): 26.2 (#260), Phi-4: 32.6 (#209)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| GPQA Diamond | 47.3% | 56.1% |
| LMArena Expert | 1233 | 1203 |
| MMLU | 77.9% | 84.8% |
| MMLU-Pro | 67.8% | — |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| GPQA (HELM) | 43.7% | — |
| BoolQ | 85.8% | — |
Multimodal Not comparable
Gemini 1.5 Flash (May 2024): 36.0 (#81), Phi-4: —
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| LMArena Vision | 1141 | — |
| Video-MME | 70.3% | — |
| GeoBench | 76% | — |
Multilingual Gemini 1.5 Flash (May 2024) leads
Gemini 1.5 Flash (May 2024): 42.9 (#189), Phi-4: 37.2 (#237)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| LMArena Non-English | 1278 | 1197 |
| LMArena Chinese | 1295 | 1212 |
| LMArena French | 1258 | 1224 |
| LMArena German | 1262 | 1222 |
| LMArena Japanese | 1252 | 1158 |
| LMArena Korean | 1221 | 1151 |
| LMArena Russian | 1288 | 1209 |
| LMArena Spanish | 1243 | 1234 |
Instruction Following Gemini 1.5 Flash (May 2024) leads
Gemini 1.5 Flash (May 2024): 66.8 (#205), Phi-4: 60.4 (#251)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1258 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
| IFEval | 83.1% | — |
Long Context Gemini 1.5 Flash (May 2024) leads
Gemini 1.5 Flash (May 2024): 39.0 (#187), Phi-4: 36.9 (#226)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| LMArena Longer Query | 1284 | 1217 |
Writing & Preference Gemini 1.5 Flash (May 2024) leads
Gemini 1.5 Flash (May 2024): 48.7 (#196), Phi-4: 40.5 (#244)
| Benchmark | Gemini 1.5 Flash (May 2024) | Phi-4 |
|---|---|---|
| LMArena Text | 1287 | 1217 |
| LMArena Creative Writing | 1285 | 1182 |
| LMArena Multi-Turn | 1253 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| WildBench | 79.2% | — |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Gemini 1.5 Flash (May 2024) better than Phi-4?
Gemini 1.5 Flash (May 2024) is the stronger model overall, scoring 33.2 to 31.2 on the Noometry Index.
Is Gemini 1.5 Flash (May 2024) or Phi-4 better for coding?
They score almost the same on coding (34.4 vs 34.4); test both on your own repository before choosing.
How many benchmarks do Gemini 1.5 Flash (May 2024) and Phi-4 share?
25 benchmarks have published results for both models. Gemini 1.5 Flash (May 2024) has 42 scored results on Noometry and Phi-4 has 37.