# Qwen1.5-72B vs Qwen3-4B

> Qwen3-4B is the stronger model overall, scoring 31.9 to 30.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen1-5-72b-vs-qwen3-4b
- Last updated: 2026-10-12
- Shared benchmarks: 1

## Summary

- They share 1 benchmark with published results for both. Qwen1.5-72B scores higher in 2 categories and Qwen3-4B in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen3-4B leads 33.0 to 11.5.
- The biggest single-benchmark swing is GPQA Diamond: 28.8% for Qwen1.5-72B and 52.3% for Qwen3-4B.

## Snapshot

| | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 30.8 | 31.9 |
| Rank | 285 | 264 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Qwen1.5-72B: 31.9 (#277)
- Qwen3-4B: —

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| BigCodeBench Instruct | 33.2% | — |
| LMArena Coding | 1165 | — |
| BigCodeBench Complete | 40.3% | — |
| HumanEval+ | 59.1% | — |
| MBPP+ | 61.6% | — |

## Agentic & Tool Use

- Qwen1.5-72B: —
- Qwen3-4B: 27.6 (#100)

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 35.7% |

## Reasoning

- Qwen1.5-72B: 22.2 (#203)
- Qwen3-4B: 19.2 (#268)

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1148 | — |

## Math

- Qwen1.5-72B: 33.2 (#205)
- Qwen3-4B: 29.7 (#240)

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| MathArena Final-Answer Competitions | — | 38.5% |
| OTIS Mock AIME 2024-2025 | — | 52.2% |
| LMArena Math | 1164 | — |

## Knowledge

- Qwen1.5-72B: 11.5 (#300)
- Qwen3-4B: 33.0 (#208)

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| GPQA Diamond | 28.8% | 52.3% |
| Vectara Hallucination Rate | — | 5.7% |
| LMArena Expert | 1136 | — |

## Multilingual

- Qwen1.5-72B: 33.2 (#253)
- Qwen3-4B: —

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1135 | — |
| LMArena Chinese | 1186 | — |
| LMArena French | 1159 | — |
| LMArena German | 1084 | — |
| LMArena Japanese | 1061 | — |
| LMArena Korean | 1050 | — |
| LMArena Russian | 1104 | — |
| LMArena Spanish | 1110 | — |

## Instruction Following

- Qwen1.5-72B: 59.3 (#256)
- Qwen3-4B: —

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1141 | — |

## Long Context

- Qwen1.5-72B: 35.1 (#243)
- Qwen3-4B: —

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| LMArena Longer Query | 1157 | — |

## Writing & Preference

- Qwen1.5-72B: 37.3 (#258)
- Qwen3-4B: —

| Benchmark | Qwen1.5-72B | Qwen3-4B |
|---|---|---|
| LMArena Text | 1166 | — |
| LMArena Creative Writing | 1137 | — |
| LMArena Multi-Turn | 1160 | — |

## FAQ

### Is Qwen1.5-72B better than Qwen3-4B?

Qwen3-4B is the stronger model overall, scoring 31.9 to 30.8 on the Noometry Index.

### How many benchmarks do Qwen1.5-72B and Qwen3-4B share?

1 benchmark has published results for both models. Qwen1.5-72B has 22 scored results on Noometry and Qwen3-4B has 6.