# GPT-4 vs Qwen3.8 Max

> Qwen3.8 Max is the stronger model overall, scoring 56.8 to 29.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-vs-qwen3-8-max
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Qwen3.8 Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 10.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.1% for GPT-4 and 100% for Qwen3.8 Max.
- Qwen3.8 Max is cheaper at $2 / $6 per million input/output tokens, against $30 / $60 for GPT-4.
- Qwen3.8 Max accepts more context: 1M tokens versus 8K.

## Snapshot

| | GPT-4 | Qwen3.8 Max |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 29.1 | 56.8 |
| Rank | 316 | 22 |
| Context | 8K | 1M |
| Input $/M | $30 | $2 |
| Output $/M | $60 | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4: 31.6 (#283)
- Qwen3.8 Max: 53.5 (#29)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1254 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |

## Agentic & Tool Use

- GPT-4: —
- Qwen3.8 Max: 45.4 (#14)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| METR Time Horizons | 36.1% | — |

## Reasoning

- GPT-4: 17.8 (#289)
- Qwen3.8 Max: 54.4 (#26)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| Chess Puzzles | 4% | 40% |
| LMArena Hard Prompts | 1241 | 1496 |
| Mystery Game Puzzles | 12% | 38% |
| DTBench | 62.7% | 92% |
| LMCA | 17.1% | 46.2% |
| Epoch Capabilities Index | 125.89 | 156.41 |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| BIG-Bench Hard | 75.1% | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |

## Math

- GPT-4: 10.8 (#309)
- Qwen3.8 Max: 73.2 (#20)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 100% |
| LMArena Math | 1269 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |

## Knowledge

- GPT-4: 18.4 (#282)
- Qwen3.8 Max: 61.7 (#27)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 35.7% | 92.7% |
| LMArena Expert | 1211 | 1507 |
| SimpleQA Verified | — | 47.3% |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |

## Multimodal

- GPT-4: —
- Qwen3.8 Max: 37.2 (#75)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |

## Multilingual

- GPT-4: 40.6 (#215)
- Qwen3.8 Max: 56.7 (#18)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1246 | 1472 |
| LMArena Chinese | 1242 | 1538 |
| LMArena French | 1283 | 1503 |
| LMArena German | 1251 | 1483 |
| LMArena Japanese | 1209 | 1467 |
| LMArena Korean | 1184 | 1461 |
| LMArena Russian | 1251 | 1481 |
| LMArena Spanish | 1261 | 1492 |

## Instruction Following

- GPT-4: 65.3 (#222)
- Qwen3.8 Max: 77.6 (#17)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1241 | 1479 |

## Long Context

- GPT-4: 37.7 (#212)
- Qwen3.8 Max: 45.6 (#31)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1244 | 1489 |

## Writing & Preference

- GPT-4: 34.9 (#268)
- Qwen3.8 Max: 67.1 (#30)

| Benchmark | GPT-4 | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1263 | 1483 |
| LMArena Creative Writing | 1244 | 1479 |
| LMArena Multi-Turn | 1257 | 1489 |
| EQ-Bench Creative Writing | 752 | — |

## FAQ

### Is GPT-4 better than Qwen3.8 Max?

Qwen3.8 Max is the stronger model overall, scoring 56.8 to 29.1 on the Noometry Index.

### Which is cheaper, GPT-4 or Qwen3.8 Max?

Qwen3.8 Max is cheaper. It lists at $2 per million input tokens and $6 per million output tokens; GPT-4 lists at $30 and $60.

### Is GPT-4 or Qwen3.8 Max better for coding?

Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 31.6 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.8 Max does, with 1M tokens against 8K.

### How many benchmarks do GPT-4 and Qwen3.8 Max share?

24 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Qwen3.8 Max has 39.