GeekRadar / The comparison desk

COMPARE.

AI models

alibaba

Qwen3.6 Max Preview

AI model Released Apr 20, 2026

Context window: 262,144 tokens
Maximum output: 65,536 tokens
Open weights: No

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

Qwen3.6 Max Preview 1446.7 Qwen3.6-Max-Preview · 95% interval 1438.6–1454.9 · 5,419 votes · 2026-10-28

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

Qwen3.6 Max Preview

Tested configuration Radar task score

Unrated

2/4 core tests · not enough evidence

3 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

Qwen3.6 Max Preview 87.4^*% Approx. 95% interval: 82.7–92.0% · 2026-08-07

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

Qwen3.6 Max Preview Not available

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

Qwen3.6 Max Preview Not available

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

Qwen3.6 Max Preview 52.0^*% Approx. 95% interval: 48.9–55.1% · 2026-08-27

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

Qwen3.6 Max Preview 91.1^*% Approx. 95% interval: 82.7–99.5% · 2026-08-07

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Test version

Arena text preference / chinesehuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,489.15 rating

Confidence interval: 1,458.16 – 1,520.13

356 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,471.12 rating

Confidence interval: 1,456.03 – 1,486.21

1613 votes

Oct 28, 2026

Category: coding
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,434.66 rating

Confidence interval: 1,412.30 – 1,457.02

756 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,451.32 rating

Confidence interval: 1,439.38 – 1,463.27

2471 votes

Oct 28, 2026

Category: english
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,441.89 rating

Confidence interval: 1,431.03 – 1,452.75

4088 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,476.65 rating

Confidence interval: 1,451.68 – 1,501.62

555 votes

Oct 28, 2026

Category: expert
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,450.94 rating

Confidence interval: 1,409.39 – 1,492.49

214 votes

Oct 28, 2026

Category: french
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,456.59 rating

Confidence interval: 1,446.43 – 1,466.74

3568 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,456.82 rating

Confidence interval: 1,442.29 – 1,471.34

1712 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,427.49 rating

Confidence interval: 1,409.26 – 1,445.73

1072 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,422.77 rating

Confidence interval: 1,403.73 – 1,441.81

1004 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,453.22 rating

Confidence interval: 1,423.34 – 1,483.11

402 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,466.09 rating

Confidence interval: 1,446.84 – 1,485.35

930 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,475.11 rating

Confidence interval: 1,445.13 – 1,505.09

365 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,459.42 rating

Confidence interval: 1,428.28 – 1,490.55

369 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,465.76 rating

Confidence interval: 1,453.10 – 1,478.42

2257 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,443.27 rating

Confidence interval: 1,425.71 – 1,460.84

1188 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,439.11 rating

Confidence interval: 1,425.40 – 1,452.82

1891 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,457.24 rating

Confidence interval: 1,444.26 – 1,470.22

2297 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,464.46 rating

Confidence interval: 1,435.50 – 1,493.43

379 votes

Oct 28, 2026

Category: math
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,454.58 rating

Confidence interval: 1,434.68 – 1,474.47

929 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,436.55 rating

Confidence interval: 1,425.64 – 1,447.47

2948 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,446.74 rating

Confidence interval: 1,438.58 – 1,454.91

5419 votes

Oct 28, 2026

Category: overall
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,445.03 rating

Confidence interval: 1,420.22 – 1,469.84

583 votes

Oct 28, 2026

Category: russian
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-28 · rating

Qwen3.6 Max Preview

1,453.42 rating

Confidence interval: 1,416.83 – 1,490.02

252 votes

Oct 28, 2026

Category: spanish
Evaluated configuration: Qwen3.6-Max-Preview
Evaluation type: human preference

Arena text preference / chinesehuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,487.95 rating

Confidence interval: 1,456.99 – 1,518.92

357 votes

Oct 15, 2026

Category: chinese
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,471.60 rating

Confidence interval: 1,456.52 – 1,486.68

1617 votes

Oct 15, 2026

Category: coding
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,435.21 rating

Confidence interval: 1,412.86 – 1,457.56

757 votes

Oct 15, 2026

Category: creative_writing
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,451.67 rating

Confidence interval: 1,439.74 – 1,463.60

2475 votes

Oct 15, 2026

Category: english
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,442.50 rating

Confidence interval: 1,431.64 – 1,453.36

4093 votes

Oct 15, 2026

Category: exclude_ties
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,478.06 rating

Confidence interval: 1,453.08 – 1,503.04

555 votes

Oct 15, 2026

Category: expert
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,450.84 rating

Confidence interval: 1,409.27 – 1,492.41

214 votes

Oct 15, 2026

Category: french
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,457.24 rating

Confidence interval: 1,447.09 – 1,467.38

3575 votes

Oct 15, 2026

Category: hard_prompts
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,457.26 rating

Confidence interval: 1,442.75 – 1,471.77

1715 votes

Oct 15, 2026

Category: hard_prompts_english
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,428.17 rating

Confidence interval: 1,409.92 – 1,446.43

1071 votes

Oct 15, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,422.69 rating

Confidence interval: 1,403.67 – 1,441.71

1008 votes

Oct 15, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,455.25 rating

Confidence interval: 1,425.33 – 1,485.17

400 votes

Oct 15, 2026

Category: industry_legal_and_government
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,466.12 rating

Confidence interval: 1,446.88 – 1,485.36

931 votes

Oct 15, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,475.09 rating

Confidence interval: 1,445.02 – 1,505.17

364 votes

Oct 15, 2026

Category: industry_mathematical
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,459.43 rating

Confidence interval: 1,428.37 – 1,490.49

370 votes

Oct 15, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,466.51 rating

Confidence interval: 1,453.87 – 1,479.16

2261 votes

Oct 15, 2026

Category: industry_software_and_it_services
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,443.20 rating

Confidence interval: 1,425.65 – 1,460.75

1192 votes

Oct 15, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,439.45 rating

Confidence interval: 1,425.75 – 1,453.16

1895 votes

Oct 15, 2026

Category: instruction_following
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,458.12 rating

Confidence interval: 1,445.15 – 1,471.09

2301 votes

Oct 15, 2026

Category: longer_query
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,465.50 rating

Confidence interval: 1,436.47 – 1,494.53

379 votes

Oct 15, 2026

Category: math
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,455.76 rating

Confidence interval: 1,435.88 – 1,475.64

932 votes

Oct 15, 2026

Category: multi_turn
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,437.11 rating

Confidence interval: 1,426.20 – 1,448.03

2951 votes

Oct 15, 2026

Category: non_english
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,447.14 rating

Confidence interval: 1,438.98 – 1,455.30

5426 votes

Oct 15, 2026

Category: overall
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,446.69 rating

Confidence interval: 1,421.85 – 1,471.53

583 votes

Oct 15, 2026

Category: russian
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-15 · rating

Qwen3.6 Max Preview

1,453.31 rating

Confidence interval: 1,416.71 – 1,489.91

252 votes

Oct 15, 2026

Category: spanish
Evaluated configuration: qwen3.6-max-preview
Evaluation type: human preference

Run it locally

Inference · memory estimate

Qwen3.6 Max Preview

Closed weights. No supported local deployment or GPU requirement is published.

API cost

USD / month

Input tokens / monthOutput tokens / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

Qwen3.6 Max Preview

Provider offer

Capabilities

Context window: 262,144 tokens
Maximum output: 65,536 tokens
Reasoning: Yes
Tool calling: Yes
Open weights: No
Knowledge cutoff: 2025-04
Model license: Proprietary

Input modalities

text

Output modalities

text

* Provisional data