Model comparison
Phi-4 vs Qwen1.5 4b Chat
Phi-4 is the stronger model overall, scoring 30.9 to 28.3 on the Noometry Index.
Last verified . 13 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Phi-4 scores higher in 6 categories and Qwen1.5 4b Chat in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Phi-4 leads 40.7 to 23.9.
Side by side
| Phi-4 | Qwen1.5 4b Chat | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 30.9 | 28.3 |
| Released | 2024-12-11 | — |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.07 | — |
| Output $ / M tokens | $0.14 | — |
| Results tracked | 37 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Phi-4 leads
Phi-4: 33.8 (#237), Qwen1.5 4b Chat: 28.6 (#305)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Coding | 1231 | 999 |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |
Agentic & Tool Use Not comparable
Phi-4: 22.6 (#120), Qwen1.5 4b Chat: —
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |
Reasoning Too close to call
Phi-4: 17.0 (#287), Qwen1.5 4b Chat: 17.7 (#277)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Hard Prompts | 1220 | 976 |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 47.8% | — |
| LiveBench Data Analysis | 45.2% | — |
| Epoch Capabilities Index | 130.43 | — |
| LiveBench | 41.6% | — |
Math Qwen1.5 4b Chat leads
Phi-4: 19.6 (#279), Qwen1.5 4b Chat: 27.9 (#233)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Math | 1246 | 1026 |
| OTIS Mock AIME 2024-2025 | 13.8% | — |
| LiveBench Math | 42% | — |
| MATH Level 5 | 64.9% | — |
Knowledge Phi-4 leads
Phi-4: 32.5 (#204), Qwen1.5 4b Chat: 26.2 (#252)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Expert | 1203 | 980 |
| GPQA Diamond | 56.1% | — |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| MMLU | 84.8% | — |
Multilingual Phi-4 leads
Phi-4: 37.4 (#232), Qwen1.5 4b Chat: 24.2 (#285)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Non-English | 1197 | 979 |
| LMArena Chinese | 1211 | 1024 |
| LMArena German | 1222 | 902 |
| LMArena Russian | 1209 | 952 |
| LMArena French | 1223 | — |
| LMArena Japanese | 1158 | — |
| LMArena Korean | 1151 | — |
| LMArena Spanish | 1234 | — |
Instruction Following Phi-4 leads
Phi-4: 60.6 (#246), Qwen1.5 4b Chat: 49.3 (#295)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Instruction Following | 1201 | 979 |
| LiveBench Instruction Following | 58.4% | — |
Long Context Phi-4 leads
Phi-4: 36.9 (#221), Qwen1.5 4b Chat: 30.1 (#285)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Longer Query | 1217 | 988 |
Writing & Preference Phi-4 leads
Phi-4: 40.7 (#239), Qwen1.5 4b Chat: 23.9 (#304)
| Benchmark | Phi-4 | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Text | 1217 | 997 |
| LMArena Creative Writing | 1182 | 969 |
| LMArena Multi-Turn | 1206 | 977 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |
Frequently asked questions
Is Phi-4 better than Qwen1.5 4b Chat?
Phi-4 is the stronger model overall, scoring 30.9 to 28.3 on the Noometry Index.
Is Phi-4 or Qwen1.5 4b Chat better for coding?
Phi-4 scores higher on coding benchmarks: 33.8 versus 28.6 in the Noometry coding category.
How many benchmarks do Phi-4 and Qwen1.5 4b Chat share?
13 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Qwen1.5 4b Chat has 13.