# Phi-4 vs Qwen Max

> Qwen Max is the stronger model overall, scoring 34.7 to 31.2 on the Noometry Index. Phi-4 costs 32× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/phi-4-vs-qwen-max
- Last updated: 2026-10-11
- Shared benchmarks: 20

## Summary

- They share 20 benchmarks with published results for both. Phi-4 scores higher in 2 categories and Qwen Max in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen Max leads 25.1 to 17.7.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $1.60 / $6.40 for Qwen Max.
- Phi-4 accepts more context: 128K tokens versus 33K.
- Phi-4 has downloadable open weights; the other is API-only.

## Snapshot

| | Phi-4 | Qwen Max |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 31.2 | 34.7 |
| Rank | 279 | 230 |
| Context | 128K | 33K |
| Input $/M | $0.07 | $1.60 |
| Output $/M | $0.14 | $6.40 |
| Weights | Open | Proprietary |

## Coding

- Phi-4: 34.4 (#239)
- Qwen Max: 30.7 (#292)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| LMArena Coding | 1231 | 1288 |
| Aider Polyglot | — | 21.8% |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| BigCodeBench Complete | 55.4% | — |

## Agentic & Tool Use

- Phi-4: 22.8 (#128)
- Qwen Max: —

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| BALROG | 11.6% | — |

## Reasoning

- Phi-4: 17.7 (#291)
- Qwen Max: 25.1 (#151)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| LMArena Hard Prompts | 1220 | 1269 |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 47.8% | — |
| LiveBench Data Analysis | 45.2% | — |
| Epoch Capabilities Index | 130.42 | — |
| LiveBench | 41.6% | — |

## Math

- Phi-4: 20.8 (#285)
- Qwen Max: 22.3 (#276)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 13.8% | 16.1% |
| LMArena Math | 1246 | 1275 |
| MATH Level 5 | 64.9% | 67.2% |
| LiveBench Math | 42% | — |
| FrontierMath (Feb 2025 set) | — | 1% |

## Knowledge

- Phi-4: 32.6 (#209)
- Qwen Max: 30.3 (#228)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| GPQA Diamond | 56.1% | 56.1% |
| LMArena Expert | 1203 | 1248 |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| MMLU | 84.8% | — |

## Multilingual

- Phi-4: 37.2 (#237)
- Qwen Max: 41.8 (#202)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| LMArena Non-English | 1197 | 1263 |
| LMArena Chinese | 1212 | 1254 |
| LMArena French | 1224 | 1330 |
| LMArena German | 1222 | 1254 |
| LMArena Japanese | 1158 | 1205 |
| LMArena Korean | 1151 | 1142 |
| LMArena Russian | 1209 | 1274 |
| LMArena Spanish | 1234 | 1290 |

## Instruction Following

- Phi-4: 60.4 (#251)
- Qwen Max: 66.5 (#208)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| LMArena Instruction Following | 1201 | 1262 |
| LiveBench Instruction Following | 58.4% | — |

## Long Context

- Phi-4: 36.9 (#226)
- Qwen Max: 39.4 (#180)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| LMArena Longer Query | 1217 | 1288 |
| Fiction.LiveBench | — | 66.7% |

## Writing & Preference

- Phi-4: 40.5 (#244)
- Qwen Max: 47.8 (#205)

| Benchmark | Phi-4 | Qwen Max |
|---|---|---|
| LMArena Text | 1217 | 1282 |
| LMArena Creative Writing | 1182 | 1248 |
| LMArena Multi-Turn | 1206 | 1277 |
| Short-Story Creative Writing | 62.6% | — |
| LiveBench Language | 25.6% | — |

## FAQ

### Is Phi-4 better than Qwen Max?

Qwen Max is the stronger model overall, scoring 34.7 to 31.2 on the Noometry Index. Phi-4 costs 32× less per token, which makes it the better buy when Qwen Max's lead doesn't matter for your workload.

### Which is cheaper, Phi-4 or Qwen Max?

Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Qwen Max lists at $1.60 and $6.40.

### Is Phi-4 or Qwen Max better for coding?

Phi-4 scores higher on coding benchmarks: 34.4 versus 30.7 in the Noometry coding category.

### Which has the bigger context window?

Phi-4 does, with 128K tokens against 33K.

### How many benchmarks do Phi-4 and Qwen Max share?

20 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Qwen Max has 23.