# GPT-5 vs Qwen3.6 Plus

> GPT-5 is the stronger model overall, scoring 50.9 to 47.5 on the Noometry Index. Qwen3.6 Plus costs 3.1× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-vs-qwen3-6-plus
- Last updated: 2026-10-11
- Shared benchmarks: 33

## Summary

- They share 33 benchmarks with published results for both. GPT-5 scores higher in 6 categories and Qwen3.6 Plus in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in long context, where GPT-5 leads 69.5 to 45.2.
- The biggest single-benchmark swing is Chess Puzzles: 37% for GPT-5 and 17% for Qwen3.6 Plus.
- Qwen3.6 Plus is cheaper at $0.50 / $3 per million input/output tokens, against $1.25 / $10 for GPT-5.
- Qwen3.6 Plus accepts more context: 1M tokens versus 400K.

## Snapshot

| | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 50.9 | 47.5 |
| Rank | 45 | 62 |
| Context | 400K | 1M |
| Input $/M | $1.25 | $0.50 |
| Output $/M | $10 | $3 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5: 50.3 (#47)
- Qwen3.6 Plus: 40.8 (#130)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| SWE-bench Verified | 73.6% | 57.9% |
| LMArena WebDev | 1418 | 1461 |
| SciCode | 42.9% | 40.7% |
| LMArena Coding | 1436 | 1467 |
| ALE-Bench | 1,162 | 670.15 |
| SWE-bench Verified (bash only) | 65% | — |
| Aider Polyglot | 88% | — |
| GSO | 6.9% | — |
| WeirdML | 60.7% | — |
| AlgoTune | 1.67 | — |

## Agentic & Tool Use

- GPT-5: 33.1 (#56)
- Qwen3.6 Plus: —

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| Terminal-Bench | 49.6% | — |
| GDPval | 34.8% | — |
| Remote Labor Index | 1.7% | — |
| DeepResearch Bench | 49.6% | — |
| BALROG | 32.8% | — |
| LMArena Search | 1133 | — |
| METR Time Horizons | 69.6% | — |
| Vending-Bench 2 | — | 5,115 |

## Reasoning

- GPT-5: 38.3 (#64)
- Qwen3.6 Plus: 29.3 (#93)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| CritPt | 12.6% | 2.9% |
| Chess Puzzles | 37% | 17% |
| LMArena Hard Prompts | 1416 | 1449 |
| Mystery Game Puzzles | 23% | 12% |
| DTBench | 90.7% | 81.9% |
| LMCA | 40% | 33.1% |
| Epoch Capabilities Index | 150 | 147.65 |
| ARC-AGI-2 | 9.9% | — |
| SimpleBench | 56.7% | — |
| Kagi LLM Benchmark | 72.7% | — |
| NYT Connections (extended) | — | 60.3% |
| ARC-AGI-1 | 65.7% | — |
| EnigmaEval | 10.5% | — |
| Thematic Generalization | — | 59.5% |
| EBR-Bench | 12.7% | — |
| ForecastBench | 61.4 | — |

## Math

- GPT-5: 55.0 (#44)
- Qwen3.6 Plus: 51.8 (#54)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| FrontierMath (Tiers 1-3) | 55.4% | 38.2% |
| OTIS Mock AIME 2024-2025 | 91.4% | 93.3% |
| LMArena Math | 1407 | 1450 |
| FrontierMath (Feb 2025 set) | 32.4% | 26.2% |
| FrontierMath Tier 4 (v1) | 12.5% | 8.3% |
| FrontierMath Tier 4 | 22% | — |
| ProofBench | 18% | — |
| Omni-MATH | 64.7% | — |
| MATH Level 5 | 98.1% | — |

## Knowledge

- GPT-5: 56.6 (#43)
- Qwen3.6 Plus: 56.1 (#45)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| GPQA Diamond | 86.2% | 88.4% |
| SimpleQA Verified | 50.1% | 44.1% |
| LMArena Expert | 1419 | 1454 |
| Humanity's Last Exam | 25.3% | — |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.3% | — |
| Vectara Hallucination Rate | 14.7% | — |
| GPQA (HELM) | 79.2% | — |

## Multimodal

- GPT-5: 46.8 (#13)
- Qwen3.6 Plus: —

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Vision | 1232 | — |
| GeoBench | 81% | — |
| VPCT | 66% | — |

## Multilingual

- GPT-5: 51.4 (#110)
- Qwen3.6 Plus: 53.3 (#70)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Non-English | 1397 | 1424 |
| LMArena Chinese | 1422 | 1477 |
| LMArena French | 1410 | 1455 |
| LMArena German | 1416 | 1452 |
| LMArena Japanese | 1409 | 1389 |
| LMArena Korean | 1360 | 1379 |
| LMArena Russian | 1406 | 1434 |
| LMArena Spanish | 1399 | 1432 |

## Instruction Following

- GPT-5: 73.8 (#113)
- Qwen3.6 Plus: 75.0 (#74)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Instruction Following | 1388 | 1425 |
| IFEval | 87.5% | — |

## Long Context

- GPT-5: 69.5 (#2)
- Qwen3.6 Plus: 45.2 (#49)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Longer Query | 1399 | 1439 |
| Fiction.LiveBench | 97.2% | — |
| CL-bench | — | 20.3% |

## Writing & Preference

- GPT-5: 63.4 (#65)
- Qwen3.6 Plus: 62.2 (#82)

| Benchmark | GPT-5 | Qwen3.6 Plus |
|---|---|---|
| LMArena Text | 1406 | 1437 |
| LMArena Creative Writing | 1365 | 1404 |
| LMArena Multi-Turn | 1426 | 1438 |
| Short-Story Creative Writing | 86% | — |
| EQ-Bench Creative Writing | 1627 | — |
| WildBench | 85.7% | — |

## FAQ

### Is GPT-5 better than Qwen3.6 Plus?

GPT-5 is the stronger model overall, scoring 50.9 to 47.5 on the Noometry Index. Qwen3.6 Plus costs 3.1× less per token, which makes it the better buy when GPT-5's lead doesn't matter for your workload.

### Which is cheaper, GPT-5 or Qwen3.6 Plus?

Qwen3.6 Plus is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; GPT-5 lists at $1.25 and $10.

### Is GPT-5 or Qwen3.6 Plus better for coding?

GPT-5 scores higher on coding benchmarks: 50.3 versus 40.8 in the Noometry coding category.

### Which has the bigger context window?

Qwen3.6 Plus does, with 1M tokens against 400K.

### How many benchmarks do GPT-5 and Qwen3.6 Plus share?

33 benchmarks have published results for both models. GPT-5 has 69 scored results on Noometry and Qwen3.6 Plus has 37.