Decision support
Qwen3.6 Max Preview vs GPT-5.5 Instant
Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.
Benchmark results
Head to head on 3 benchmarks both models report
- Qwen3.6 Max Preview2leads
- GPT-5.5 Instant1lead
Arena Text
Arena rating · Higher is better
- Qwen3.6 Max Preview1460Arena #62CI 1451 to 1468
- GPT-5.5 InstantBest1473+14 ahead · Arena #44CI 1468 to 1478
Runs and sources
Qwen3.6 Max Preview- default146095% CI 1451 to 1468 · 5,386 votes
GPT-5.5 Instant- default147395% CI 1468 to 1478 · 26,865 votes
Arena WebDev
Arena rating · Higher is better
- Qwen3.6 Max Preview1482Arena #75CI 1470 to 1495
- GPT-5.5 InstantNot reported
Runs and sources
Qwen3.6 Max Preview- default148295% CI 1470 to 1495 · 2,674 votes
GPT-5.5 InstantNot reported
Arena Vision
Arena rating · Higher is better
- Qwen3.6 Max PreviewNot reported
- GPT-5.5 Instant1277Arena #33CI 1269 to 1286
Runs and sources
Qwen3.6 Max PreviewNot reported
GPT-5.5 Instant- default127795% CI 1269 to 1286 · 7,567 votes
Arena Document
Arena rating · Higher is better
- Qwen3.6 Max PreviewNot reported
- GPT-5.5 Instant1460Arena #26CI 1452 to 1469
Runs and sources
Qwen3.6 Max PreviewNot reported
GPT-5.5 Instant- default146095% CI 1452 to 1469 · 8,834 votes
FrontierMath-2025-02-28-Private
Best published score · Higher is better
- Qwen3.6 Max Preview23.1%#20 of 58
- GPT-5.5 InstantNot reported
Runs and sources
Qwen3.6 Max Preview- default23.1%
GPT-5.5 InstantNot reported
Source:Epoch AI evaluationFrontierMath-Tier-4-2025-07-01-Private
Best published score · Higher is better
- Qwen3.6 Max Preview4.2%#18 of 46
- GPT-5.5 InstantNot reported
Runs and sources
Qwen3.6 Max Preview- default4.2%
GPT-5.5 InstantNot reported
Source:Epoch AI evaluationFrontierMath-Tier-4-v2-Private
Best published score · Higher is better
- Qwen3.6 Max PreviewNot reported
- GPT-5.5 Instant2.4%#64 of 72
Runs and sources
Qwen3.6 Max PreviewNot reported
GPT-5.5 Instant- default2.4%
Source:Epoch AI evaluationFrontierMath-Tiers-1-3-v2-Private
Best published score · Higher is better
- Qwen3.6 Max PreviewNot reported
- GPT-5.5 Instant26.3%#71 of 87
Runs and sources
Qwen3.6 Max PreviewNot reported
GPT-5.5 Instant- default26.3%
Source:Epoch AI evaluationGPQA diamond
Best published score · Higher is better
- Qwen3.6 Max PreviewBest87.4%+4.9 pts ahead · #59 of 161
- GPT-5.5 Instant82.5%#80 of 161
Runs and sources
Qwen3.6 Max Preview- default87.4%
GPT-5.5 Instant- default82.5%
Source:Epoch AI evaluationOTIS Mock AIME 2024-2025
Best published score · Higher is better
- Qwen3.6 Max PreviewBest91.1%+23.1 pts ahead · #54 of 158
- GPT-5.5 Instant68.1%#101 of 158
Runs and sources
Qwen3.6 Max Preview- default91.1%
GPT-5.5 Instant- default68.1%
Source:Epoch AI evaluationSimpleQA Verified
Best published score · Higher is better
- Qwen3.6 Max Preview52.0%#26 of 87
- GPT-5.5 InstantNot reported
Runs and sources
Qwen3.6 Max Preview- default52.0%
GPT-5.5 InstantNot reported
Source:Epoch AI evaluationSWE-bench Verified
Best published score · Higher is better
- Qwen3.6 Max Preview76.7%#7 of 35
- GPT-5.5 InstantNot reported
Runs and sources
Qwen3.6 Max Preview- default76.7%
GPT-5.5 InstantNot reported
Source:Epoch AI evaluation
Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.
Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.
Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.
Model specifications
API pricing · USD per 1M tokens
Capabilities and modalities
Architecture and availability
A dash means the field is not reported. Provider-specific performance and pricing may differ.