# GPT-4 vs Qwen3-4B

> Qwen3-4B is the stronger model overall, scoring 31.9 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-vs-qwen3-4b
- Last updated: 2026-10-10
- Shared benchmarks: 3

## Summary

- They share 3 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Qwen3-4B in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3-4B leads 29.7 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 52.2% for Qwen3-4B.
- Qwen3-4B has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4 | Qwen3-4B |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 29.1 | 31.9 |
| Rank | 316 | 264 |
| Context | 8K | — |
| Input $/M | $30 | — |
| Output $/M | $60 | — |
| Weights | Proprietary | Open |

## Coding

- GPT-4: 31.6 (#283)
- Qwen3-4B: —

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| LMArena Coding | 1254 | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |

## Agentic & Tool Use

- GPT-4: —
- Qwen3-4B: 27.6 (#100)

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 35.7% |
| METR Time Horizons | 36.1% | — |

## Reasoning

- GPT-4: 17.8 (#289)
- Qwen3-4B: 19.2 (#268)

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| Chess Puzzles | 4% | 4% |
| LMArena Hard Prompts | 1241 | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| Epoch Capabilities Index | 125.89 | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |

## Math

- GPT-4: 10.8 (#309)
- Qwen3-4B: 29.7 (#240)

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 52.2% |
| MathArena Final-Answer Competitions | — | 38.5% |
| LMArena Math | 1269 | — |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |

## Knowledge

- GPT-4: 18.4 (#282)
- Qwen3-4B: 33.0 (#208)

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| GPQA Diamond | 35.7% | 52.3% |
| Vectara Hallucination Rate | — | 5.7% |
| LMArena Expert | 1211 | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |

## Multilingual

- GPT-4: 40.6 (#215)
- Qwen3-4B: —

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1246 | — |
| LMArena Chinese | 1242 | — |
| LMArena French | 1283 | — |
| LMArena German | 1251 | — |
| LMArena Japanese | 1209 | — |
| LMArena Korean | 1184 | — |
| LMArena Russian | 1251 | — |
| LMArena Spanish | 1261 | — |

## Instruction Following

- GPT-4: 65.3 (#222)
- Qwen3-4B: —

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| LMArena Instruction Following | 1241 | — |

## Long Context

- GPT-4: 37.7 (#212)
- Qwen3-4B: —

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| LMArena Longer Query | 1244 | — |

## Writing & Preference

- GPT-4: 34.9 (#268)
- Qwen3-4B: —

| Benchmark | GPT-4 | Qwen3-4B |
|---|---|---|
| LMArena Text | 1263 | — |
| LMArena Creative Writing | 1244 | — |
| EQ-Bench Creative Writing | 752 | — |
| LMArena Multi-Turn | 1257 | — |

## FAQ

### Is GPT-4 better than Qwen3-4B?

Qwen3-4B is the stronger model overall, scoring 31.9 to 29.1 on the Noometry Index.

### How many benchmarks do GPT-4 and Qwen3-4B share?

3 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Qwen3-4B has 6.