Model comparison
Phi 3 Mini 128k Instruct vs Phi-4
Phi-4 is the stronger model overall, scoring 31.2 to 29.7 on the Noometry Index.
Last verified . 19 shared benchmarks.
MicrosoftSummary
- They share 19 benchmarks with published results for both. Phi 3 Mini 128k Instruct scores higher in 2 categories and Phi-4 in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Phi-4 leads 40.5 to 27.1.
- The biggest single-benchmark swing is BigCodeBench Instruct: 29.6% for Phi 3 Mini 128k Instruct and 45.5% for Phi-4.
Side by side
| Phi 3 Mini 128k Instruct | 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 | 19 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-4 leads
Phi 3 Mini 128k Instruct: 28.8 (#312), Phi-4: 34.4 (#239)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| BigCodeBench Instruct | 29.6% | 45.5% |
| LMArena Coding | 1039 | 1231 |
| BigCodeBench Complete | 40.6% | 55.4% |
| LiveBench Coding | — | 30.7% |
Agentic & Tool Use Not comparable
Phi 3 Mini 128k Instruct: —, Phi-4: 22.8 (#128)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 28.8% |
| BALROG | — | 11.6% |
Reasoning Phi 3 Mini 128k Instruct leads
Phi 3 Mini 128k Instruct: 19.6 (#256), Phi-4: 17.7 (#291)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| LMArena Hard Prompts | 1028 | 1220 |
| Chess Puzzles | — | 1% |
| LiveBench Reasoning | — | 47.8% |
| LiveBench Data Analysis | — | 45.2% |
| Epoch Capabilities Index | — | 130.42 |
| LiveBench | — | 41.6% |
Math Phi 3 Mini 128k Instruct leads
Phi 3 Mini 128k Instruct: 31.6 (#222), Phi-4: 20.8 (#285)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| LMArena Math | 1089 | 1246 |
| OTIS Mock AIME 2024-2025 | — | 13.8% |
| LiveBench Math | — | 42% |
| MATH Level 5 | — | 64.9% |
Knowledge Phi-4 leads
Phi 3 Mini 128k Instruct: 26.8 (#254), Phi-4: 32.6 (#209)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| LMArena Expert | 984 | 1203 |
| GPQA Diamond | — | 56.1% |
| Confabulations | — | 29.4% |
| Vectara Hallucination Rate | — | 3.7% |
| MMLU | — | 84.8% |
Multilingual Phi-4 leads
Phi 3 Mini 128k Instruct: 25.2 (#285), Phi-4: 37.2 (#237)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| LMArena Non-English | 1000 | 1197 |
| LMArena Chinese | 1016 | 1212 |
| LMArena French | 1039 | 1224 |
| LMArena German | 1006 | 1222 |
| LMArena Japanese | 899 | 1158 |
| LMArena Korean | 856 | 1151 |
| LMArena Russian | 1004 | 1209 |
| LMArena Spanish | 1059 | 1234 |
Instruction Following Phi-4 leads
Phi 3 Mini 128k Instruct: 51.8 (#294), Phi-4: 60.4 (#251)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| LMArena Instruction Following | 1023 | 1201 |
| LiveBench Instruction Following | — | 58.4% |
Long Context Phi-4 leads
Phi 3 Mini 128k Instruct: 30.4 (#289), Phi-4: 36.9 (#226)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| LMArena Longer Query | 996 | 1217 |
Writing & Preference Phi-4 leads
Phi 3 Mini 128k Instruct: 27.1 (#301), Phi-4: 40.5 (#244)
| Benchmark | Phi 3 Mini 128k Instruct | Phi-4 |
|---|---|---|
| LMArena Text | 1050 | 1217 |
| LMArena Creative Writing | 1024 | 1182 |
| LMArena Multi-Turn | 989 | 1206 |
| Short-Story Creative Writing | — | 62.6% |
| LiveBench Language | — | 25.6% |
Frequently asked questions
Is Phi 3 Mini 128k Instruct better than Phi-4?
Phi-4 is the stronger model overall, scoring 31.2 to 29.7 on the Noometry Index.
Is Phi 3 Mini 128k Instruct or Phi-4 better for coding?
Phi-4 scores higher on coding benchmarks: 34.4 versus 28.8 in the Noometry coding category.
How many benchmarks do Phi 3 Mini 128k Instruct and Phi-4 share?
19 benchmarks have published results for both models. Phi 3 Mini 128k Instruct has 19 scored results on Noometry and Phi-4 has 37.