GeekRadar / The comparison desk
COMPARE.
AI models
alibaba

QwQ-32B-Preview

Model evaluation

Capability & preference

Human preference · Arena

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

QwQ-32B-Preview 1161.8 QwQ-32B-Preview · 95% interval 1150.4–1173.1 · 3,231 votes · 2026-10-28

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 · 2026-10-28 · rating

QwQ-32B-Preview

1,217.79 rating

Confidence interval: 1,177.53 – 1,258.04

206 votes

Category
chinese
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,154.94 rating

Confidence interval: 1,131.24 – 1,178.64

566 votes

Category
coding
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,133.44 rating

Confidence interval: 1,105.27 – 1,161.61

454 votes

Category
creative_writing
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,184.27 rating

Confidence interval: 1,169.79 – 1,198.76

1783 votes

Category
english
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,022.40 rating

Confidence interval: 1,004.31 – 1,040.50

2230 votes

Category
exclude_ties
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,134.28 rating

Confidence interval: 1,098.60 – 1,169.96

232 votes

Category
expert
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,162.43 rating

Confidence interval: 1,143.35 – 1,181.51

852 votes

Category
hard_prompts
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,170.89 rating

Confidence interval: 1,146.47 – 1,195.31

519 votes

Category
hard_prompts_english
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,098.29 rating

Confidence interval: 1,066.84 – 1,129.73

374 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,114.47 rating

Confidence interval: 1,088.47 – 1,140.46

540 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,143.10 rating

Confidence interval: 1,102.31 – 1,183.89

214 votes

Category
industry_legal_and_government
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,153.89 rating

Confidence interval: 1,128.87 – 1,178.92

583 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,221.43 rating

Confidence interval: 1,195.00 – 1,247.85

422 votes

Category
industry_mathematical
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,154.09 rating

Confidence interval: 1,133.82 – 1,174.35

824 votes

Category
industry_software_and_it_services
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,139.37 rating

Confidence interval: 1,118.82 – 1,159.92

855 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,155.77 rating

Confidence interval: 1,139.94 – 1,171.59

1329 votes

Category
instruction_following
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,166.46 rating

Confidence interval: 1,140.67 – 1,192.25

475 votes

Category
longer_query
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,213.46 rating

Confidence interval: 1,189.14 – 1,237.78

480 votes

Category
math
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,140.66 rating

Confidence interval: 1,113.77 – 1,167.55

487 votes

Category
multi_turn
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,129.46 rating

Confidence interval: 1,113.86 – 1,145.06

1448 votes

Category
non_english
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,161.77 rating

Confidence interval: 1,150.44 – 1,173.10

3231 votes

Category
overall
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

QwQ-32B-Preview

1,114.37 rating

Confidence interval: 1,088.30 – 1,140.44

527 votes

Category
russian
Evaluated configuration
QwQ-32B-Preview
Evaluation type
human preference

Run it locally

Inference · memory estimate

QwQ-32B-Preview

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Apache 2.0