# Phi-4 vs Qwen2.5-VL 72B Instruct

> Phi-4 is the stronger model overall, scoring 30.9 to 29.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/phi-4-vs-qwen2-5-vl-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- The widest gap is in agentic & tool use, where Phi-4 leads 22.6 to 18.4.
- Phi-4 is cheaper at $0.07 / $0.14 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
- Qwen2.5-VL 72B Instruct accepts more context: 131K tokens versus 128K.

## Snapshot

| | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Provider | Microsoft | Alibaba (Qwen) |
| Noometry Index | 30.9 | 29.4 |
| Rank | 269 | 298 |
| Context | 128K | 131K |
| Input $/M | $0.07 | $2.80 |
| Output $/M | $0.14 | $8.40 |
| Weights | Open | Open |

## Coding

- Phi-4: 33.8 (#237)
- Qwen2.5-VL 72B Instruct: —

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 30.7% | — |
| LMArena Coding | 1231 | — |
| BigCodeBench Complete | 55.4% | — |

## Agentic & Tool Use

- Phi-4: 22.6 (#120)
- Qwen2.5-VL 72B Instruct: 18.4 (#136)

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 28.8% | — |
| OSWorld | — | 5% |
| BALROG | 11.6% | — |

## Reasoning

- Phi-4: 17.0 (#287)
- Qwen2.5-VL 72B Instruct: 20.0 (#228)

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| Kagi LLM Benchmark | — | 36% |
| Chess Puzzles | 1% | — |
| LiveBench Reasoning | 47.8% | — |
| LMArena Hard Prompts | 1220 | — |
| LiveBench Data Analysis | 45.2% | — |
| Epoch Capabilities Index | 130.43 | — |
| LiveBench | 41.6% | — |

## Math

- Phi-4: 19.6 (#279)
- Qwen2.5-VL 72B Instruct: —

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 13.8% | — |
| LiveBench Math | 42% | — |
| LMArena Math | 1246 | — |
| MATH Level 5 | 64.9% | — |

## Knowledge

- Phi-4: 32.5 (#204)
- Qwen2.5-VL 72B Instruct: —

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| GPQA Diamond | 56.1% | — |
| Confabulations | 29.4% | — |
| Vectara Hallucination Rate | 3.7% | — |
| LMArena Expert | 1203 | — |
| MMLU | 84.8% | — |

## Multimodal

- Phi-4: —
- Qwen2.5-VL 72B Instruct: 32.9 (#92)

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Vision | — | 1107 |
| Video-MME | — | 73.5% |
| GeoBench | — | 62% |
| SpatialViz-Bench | — | 33.3% |

## Multilingual

- Phi-4: 37.4 (#232)
- Qwen2.5-VL 72B Instruct: —

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Non-English | 1197 | — |
| LMArena Chinese | 1211 | — |
| LMArena French | 1223 | — |
| LMArena German | 1222 | — |
| LMArena Japanese | 1158 | — |
| LMArena Korean | 1151 | — |
| LMArena Russian | 1209 | — |
| LMArena Spanish | 1234 | — |

## Instruction Following

- Phi-4: 60.6 (#246)
- Qwen2.5-VL 72B Instruct: —

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LiveBench Instruction Following | 58.4% | — |
| LMArena Instruction Following | 1201 | — |

## Long Context

- Phi-4: 36.9 (#221)
- Qwen2.5-VL 72B Instruct: —

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1217 | — |

## Writing & Preference

- Phi-4: 40.7 (#239)
- Qwen2.5-VL 72B Instruct: —

| Benchmark | Phi-4 | Qwen2.5-VL 72B Instruct |
|---|---|---|
| LMArena Text | 1217 | — |
| LMArena Creative Writing | 1182 | — |
| Short-Story Creative Writing | 62.6% | — |
| LMArena Multi-Turn | 1206 | — |
| LiveBench Language | 25.6% | — |

## FAQ

### Is Phi-4 better than Qwen2.5-VL 72B Instruct?

Phi-4 is the stronger model overall, scoring 30.9 to 29.4 on the Noometry Index.

### Which is cheaper, Phi-4 or Qwen2.5-VL 72B Instruct?

Phi-4 is cheaper. It lists at $0.07 per million input tokens and $0.14 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.

### Which has the bigger context window?

Qwen2.5-VL 72B Instruct does, with 131K tokens against 128K.

### How many benchmarks do Phi-4 and Qwen2.5-VL 72B Instruct share?

0 benchmarks have published results for both models. Phi-4 has 37 scored results on Noometry and Qwen2.5-VL 72B Instruct has 6.