Model comparison
phi-3-medium 14B vs Phi-4
Phi-4 is the stronger model overall, scoring 31.2 to 29.7 on the Noometry Index.
Last verified . 6 shared benchmarks.
MicrosoftSummary
- They share 6 benchmarks with published results for both. phi-3-medium 14B scores higher in 2 categories and Phi-4 in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi-4 leads 32.6 to 9.1.
- The biggest single-benchmark swing is MATH Level 5: 17.6% for phi-3-medium 14B and 64.9% for Phi-4.
Side by side
| phi-3-medium 14B | Phi-4 | |
|---|---|---|
| Provider | Microsoft | Microsoft |
| Noometry Index | 29.7 | 31.2 |
| Released | 2024-04-23 | 2024-12-11 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.07 |
| Output $ / M tokens | — | $0.14 |
| Results tracked | 13 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding phi-3-medium 14B leads
phi-3-medium 14B: 36.8 (#201), Phi-4: 34.4 (#239)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 37.6% | 45.5% |
| BigCodeBench Complete | 48.7% | 55.4% |
| LiveBench Coding | — | 30.7% |
| LMArena Coding | — | 1231 |
Agentic & Tool Use Not comparable
phi-3-medium 14B: —, Phi-4: 22.8 (#128)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning Not comparable
phi-3-medium 14B: —, Phi-4: 17.7 (#291)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| Epoch Capabilities Index | 121.23 | 130.42 |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LMArena Hard Prompts | — | 1220 |
| LiveBench Data Analysis | — | 45.2% |
| Adversarial NLI | 55.8% | — |
| BIG-Bench Hard | 81.4% | — |
| HellaSwag | 82.4% | — |
| LiveBench | — | 41.6% |
| WinoGrande | 81.5% | — |
Math phi-3-medium 14B leads
phi-3-medium 14B: 27.3 (#250), Phi-4: 20.8 (#285)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| MATH Level 5 | 17.6% | 64.9% |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| LMArena Math | — | 1246 |
Knowledge Phi-4 leads
phi-3-medium 14B: 9.1 (#306), Phi-4: 32.6 (#209)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| GPQA Diamond | 27.6% | 56.1% |
| MMLU | 78% | 84.8% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| LMArena Expert | — | 1203 |
| ARC (AI2) Challenge | 91.6% | — |
| OpenBookQA | 87.4% | — |
| TriviaQA | 73.9% | — |
Multilingual Not comparable
phi-3-medium 14B: —, Phi-4: 37.2 (#237)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| LMArena Non-English | — | 1197 |
| LMArena Chinese | — | 1212 |
| LMArena French | — | 1224 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1158 |
| LMArena Korean | — | 1151 |
| LMArena Russian | — | 1209 |
| LMArena Spanish | — | 1234 |
Instruction Following Not comparable
phi-3-medium 14B: —, Phi-4: 60.4 (#251)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| LiveBench Instruction Following | — | 58.4% |
| LMArena Instruction Following | — | 1201 |
Long Context Not comparable
phi-3-medium 14B: —, Phi-4: 36.9 (#226)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| LMArena Longer Query | — | 1217 |
Writing & Preference Not comparable
phi-3-medium 14B: —, Phi-4: 40.5 (#244)
| Benchmark | phi-3-medium 14B | Phi-4 |
|---|---|---|
| LMArena Text | — | 1217 |
| LMArena Creative Writing | — | 1182 |
| Short-Story Creative Writing | — | 62.6% |
| LMArena Multi-Turn | — | 1206 |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is phi-3-medium 14B better than Phi-4?
Phi-4 is the stronger model overall, scoring 31.2 to 29.7 on the Noometry Index.
Is phi-3-medium 14B or Phi-4 better for coding?
phi-3-medium 14B scores higher on coding benchmarks: 36.8 versus 34.4 in the Noometry coding category.
How many benchmarks do phi-3-medium 14B and Phi-4 share?
6 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Phi-4 has 37.