GeekRadar / The comparison desk

COMPARE.

AI models

alibaba

Qwen3 Max

AI model Released Sep 23, 2025

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 Max 1412.9 qwen3-max-2025-09-23 · 95% interval 1406.5–1419.4 · 9,027 votes · 2026-10-15

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.

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

Qwen3 Max

1,430.80 rating

Confidence interval: 1,405.90 – 1,455.71

502 votes

Oct 15, 2026

Category: chinese
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,438.71 rating

Confidence interval: 1,425.84 – 1,451.57

2030 votes

Oct 15, 2026

Category: coding
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,380.80 rating

Confidence interval: 1,363.80 – 1,397.80

1152 votes

Oct 15, 2026

Category: creative_writing
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,418.17 rating

Confidence interval: 1,408.99 – 1,427.34

3918 votes

Oct 15, 2026

Category: english
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,398.08 rating

Confidence interval: 1,388.99 – 1,407.17

6291 votes

Oct 15, 2026

Category: exclude_ties
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,397.51 rating

Confidence interval: 1,371.41 – 1,423.60

475 votes

Oct 15, 2026

Category: expert
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,437.04 rating

Confidence interval: 1,397.12 – 1,476.97

203 votes

Oct 15, 2026

Category: german
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,423.37 rating

Confidence interval: 1,414.92 – 1,431.81

4849 votes

Oct 15, 2026

Category: hard_prompts
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,428.86 rating

Confidence interval: 1,416.34 – 1,441.38

2127 votes

Oct 15, 2026

Category: hard_prompts_english
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,394.55 rating

Confidence interval: 1,380.84 – 1,408.25

1756 votes

Oct 15, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,378.95 rating

Confidence interval: 1,365.00 – 1,392.89

1660 votes

Oct 15, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,400.28 rating

Confidence interval: 1,375.84 – 1,424.72

554 votes

Oct 15, 2026

Category: industry_legal_and_government
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,413.13 rating

Confidence interval: 1,397.86 – 1,428.40

1427 votes

Oct 15, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,416.29 rating

Confidence interval: 1,390.56 – 1,442.01

489 votes

Oct 15, 2026

Category: industry_mathematical
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,405.58 rating

Confidence interval: 1,377.96 – 1,433.21

449 votes

Oct 15, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,435.94 rating

Confidence interval: 1,425.87 – 1,446.00

3322 votes

Oct 15, 2026

Category: industry_software_and_it_services
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,390.12 rating

Confidence interval: 1,377.20 – 1,403.04

1978 votes

Oct 15, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,398.01 rating

Confidence interval: 1,386.79 – 1,409.22

2659 votes

Oct 15, 2026

Category: instruction_following
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,375.78 rating

Confidence interval: 1,339.55 – 1,412.01

266 votes

Oct 15, 2026

Category: korean
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,410.38 rating

Confidence interval: 1,397.74 – 1,423.01

2091 votes

Oct 15, 2026

Category: longer_query
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,434.85 rating

Confidence interval: 1,411.14 – 1,458.56

571 votes

Oct 15, 2026

Category: math
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,434.09 rating

Confidence interval: 1,419.44 – 1,448.74

1636 votes

Oct 15, 2026

Category: multi_turn
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,400.20 rating

Confidence interval: 1,391.96 – 1,408.45

5109 votes

Oct 15, 2026

Category: non_english
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,412.93 rating

Confidence interval: 1,406.48 – 1,419.38

9027 votes

Oct 15, 2026

Category: overall
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,393.85 rating

Confidence interval: 1,364.98 – 1,422.72

419 votes

Oct 15, 2026

Category: polish
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,415.55 rating

Confidence interval: 1,394.72 – 1,436.38

730 votes

Oct 15, 2026

Category: russian
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

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

Qwen3 Max

1,420.64 rating

Confidence interval: 1,392.23 – 1,449.06

439 votes

Oct 15, 2026

Category: spanish
Evaluated configuration: qwen3-max-2025-09-23
Evaluation type: human preference

Run it locally

Inference · memory estimate

Qwen3 Max

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 Max

Provider offer

Capabilities

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

Input modalities

text

Output modalities

text