Model comparison
Phi-4 vs Qwen Plus
Qwen Plus is the stronger model overall, scoring 37.1 to 31.2 on the Noometry Index. Phi-4 costs 6.9× less per token, which makes it the better buy when Qwen Plus's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 16 benchmarks with published results for both. Phi-4 scores higher in 1 category and Qwen Plus in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen Plus leads 52.2 to 40.5.
- The biggest single-benchmark swing is GPQA Diamond: 56.1% for Phi-4 and 48.1% for Qwen Plus.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $0.40 / $1.20 for Qwen Plus.
- Qwen Plus accepts more context: 1M tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Phi-4 | Qwen Plus | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 31.2 | 37.1 |
| Released | 2024-12-11 | 2024-01-25 |
| Weights | Open | Proprietary |
| Context window | 128K | 1M |
| Max output | 4K | 33K |
| Input $ / M tokens | $0.07 | $0.40 |
| Output $ / M tokens | $0.14 | $1.20 |
| Results tracked | 37 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen Plus leads
Phi-4: 34.4 (#239), Qwen Plus: 38.9 (#167)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| LMArena Coding | 1231 | 1328 |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |
Agentic & Tool Use Not comparable
Phi-4: 22.8 (#128), Qwen Plus: —
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |
Reasoning Qwen Plus leads
Phi-4: 17.7 (#291), Qwen Plus: 28.4 (#107)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1317 |
| Kagi LLM Benchmark | — | 63.3% |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 47.8% | — |
| DTBench | — | 81.1% |
| LiveBench Data Analysis | 45.2% | — |
| LMCA | — | 24% |
| Epoch Capabilities Index | 130.42 | — |
| LiveBench | 41.6% | — |
Math Qwen Plus leads
Phi-4: 20.8 (#285), Qwen Plus: 23.3 (#271)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 13.8% | 17.8% |
| LMArena Math | 1246 | 1326 |
| MATH Level 5 | 64.9% | 65.3% |
| LiveBench Math | 42% | — |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge Phi-4 leads
Phi-4: 32.6 (#209), Qwen Plus: 27.4 (#251)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| GPQA Diamond | 56.1% | 48.1% |
| LMArena Expert | 1203 | 1328 |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| MMLU | 84.8% | — |
Multilingual Qwen Plus leads
Phi-4: 37.2 (#237), Qwen Plus: 45.1 (#175)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| LMArena Non-English | 1197 | 1310 |
| LMArena Chinese | 1212 | 1347 |
| LMArena Japanese | 1158 | 1251 |
| LMArena Russian | 1209 | 1323 |
| LMArena French | 1224 | — |
| LMArena German | 1222 | — |
| LMArena Korean | 1151 | — |
| LMArena Spanish | 1234 | — |
Instruction Following Qwen Plus leads
Phi-4: 60.4 (#251), Qwen Plus: 68.8 (#181)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| LMArena Instruction Following | 1201 | 1303 |
| LiveBench Instruction Following | 58.4% | — |
Long Context Qwen Plus leads
Phi-4: 36.9 (#226), Qwen Plus: 40.3 (#158)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| LMArena Longer Query | 1217 | 1324 |
Writing & Preference Qwen Plus leads
Phi-4: 40.5 (#244), Qwen Plus: 52.2 (#176)
| Benchmark | Phi-4 | Qwen Plus |
|---|---|---|
| LMArena Text | 1217 | 1326 |
| LMArena Creative Writing | 1182 | 1293 |
| LMArena Multi-Turn | 1206 | 1336 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |
Frequently asked questions
Is Phi-4 better than Qwen Plus?
Qwen Plus is the stronger model overall, scoring 37.1 to 31.2 on the Noometry Index. Phi-4 costs 6.9× less per token, which makes it the better buy when Qwen Plus's lead doesn't matter for your workload.
Which is cheaper, Phi-4 or Qwen Plus?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Qwen Plus lists at $0.40 and $1.20.
Is Phi-4 or Qwen Plus better for coding?
Qwen Plus scores higher on coding benchmarks: 38.9 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Qwen Plus does, with 1M tokens against 128K.
How many benchmarks do Phi-4 and Qwen Plus share?
16 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Qwen Plus has 20.