Model comparison
Phi-4 vs Qwen3 Max
Qwen3 Max is the stronger model overall, scoring 43.7 to 31.2 on the Noometry Index. Phi-4 costs 27× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Last verified . 22 shared benchmarks.
Microsoft
Alibaba (Qwen)Summary
- They share 22 benchmarks with published results for both. Phi-4 scores higher in 0 categories and Qwen3 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen3 Max leads 62.4 to 40.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 13.8% for Phi-4 and 73.3% for Qwen3 Max.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $1.20 / $6 for Qwen3 Max.
- Qwen3 Max accepts more context: 262K tokens versus 128K.
- Phi-4 has downloadable open weights; the other is API-only.
Side by side
| Phi-4 | Qwen3 Max | |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 31.2 | 43.7 |
| Released | 2024-12-11 | 2025-09-23 |
| Weights | Open | Proprietary |
| Context window | 128K | 262K |
| Max output | 4K | 66K |
| Input $ / M tokens | $0.07 | $1.20 |
| Output $ / M tokens | $0.14 | $6 |
| Results tracked | 37 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 Max leads
Phi-4: 34.4 (#239), Qwen3 Max: 43.0 (#93)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| LMArena Coding | 1231 | 1456 |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |
| ALE-Bench | — | 370.45 |
Agentic & Tool Use Not comparable
Phi-4: 22.8 (#128), Qwen3 Max: —
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |
| Vending-Bench 2 | — | 71.56 |
Reasoning Qwen3 Max leads
Phi-4: 17.7 (#291), Qwen3 Max: 22.6 (#190)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| Chess Puzzles | 1% | 4% |
| LMArena Hard Prompts | 1220 | 1448 |
| Epoch Capabilities Index | 130.42 | 142.38 |
| Kagi LLM Benchmark | — | 72.5% |
| NYT Connections (extended) | — | 30.1% |
| LiveBench Reasoning | 47.8% | — |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 82.1% |
| LiveBench Data Analysis | 45.2% | — |
| LMCA | — | 28.3% |
| LiveBench | 41.6% | — |
Math Qwen3 Max leads
Phi-4: 20.8 (#285), Qwen3 Max: 38.7 (#131)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 13.8% | 73.3% |
| LMArena Math | 1246 | 1446 |
| MATH Level 5 | 64.9% | 97.1% |
| FrontierMath (Tiers 1-3) | — | 18.9% |
| LiveBench Math | 42% | — |
Knowledge Qwen3 Max leads
Phi-4: 32.6 (#209), Qwen3 Max: 48.1 (#78)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| GPQA Diamond | 56.1% | 72.6% |
| LMArena Expert | 1203 | 1455 |
| SimpleQA Verified | — | 48.7% |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| MMLU | 84.8% | — |
Multilingual Qwen3 Max leads
Phi-4: 37.2 (#237), Qwen3 Max: 53.7 (#62)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| LMArena Non-English | 1197 | 1429 |
| LMArena Chinese | 1212 | 1478 |
| LMArena French | 1224 | 1449 |
| LMArena German | 1222 | 1463 |
| LMArena Japanese | 1158 | 1397 |
| LMArena Korean | 1151 | 1399 |
| LMArena Russian | 1209 | 1428 |
| LMArena Spanish | 1234 | 1462 |
Instruction Following Qwen3 Max leads
Phi-4: 60.4 (#251), Qwen3 Max: 74.8 (#87)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| LMArena Instruction Following | 1201 | 1419 |
| LiveBench Instruction Following | 58.4% | — |
Long Context Qwen3 Max leads
Phi-4: 36.9 (#226), Qwen3 Max: 41.6 (#134)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| LMArena Longer Query | 1217 | 1438 |
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 14.5% |
Writing & Preference Qwen3 Max leads
Phi-4: 40.5 (#244), Qwen3 Max: 62.4 (#76)
| Benchmark | Phi-4 | Qwen3 Max |
|---|---|---|
| LMArena Text | 1217 | 1439 |
| LMArena Creative Writing | 1182 | 1402 |
| LMArena Multi-Turn | 1206 | 1446 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |
Frequently asked questions
Is Phi-4 better than Qwen3 Max?
Qwen3 Max is the stronger model overall, scoring 43.7 to 31.2 on the Noometry Index. Phi-4 costs 27× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Which is cheaper, Phi-4 or Qwen3 Max?
Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Qwen3 Max lists at $1.20 and $6.
Is Phi-4 or Qwen3 Max better for coding?
Qwen3 Max scores higher on coding benchmarks: 43.0 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
Qwen3 Max does, with 262K tokens against 128K.
How many benchmarks do Phi-4 and Qwen3 Max share?
22 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Qwen3 Max has 33.