GeekRadar / The comparison desk
COMPARE.
AI models
alibaba

Qwen3.5 Flash

AI model Released

Context window
1,000,000 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.

Qwen3.5 Flash 1397.1 Qwen3.5-Flash · 95% interval 1393.4–1400.7 · 61,776 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

Qwen3.5 Flash

1,446.07 rating

Confidence interval: 1,436.06 – 1,456.09

3673 votes

Category
chinese
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,411.77 rating

Confidence interval: 1,406.14 – 1,417.40

17729 votes

Category
coding
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,343.36 rating

Confidence interval: 1,336.44 – 1,350.28

9717 votes

Category
creative_writing
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,404.41 rating

Confidence interval: 1,399.51 – 1,409.30

26633 votes

Category
english
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,376.26 rating

Confidence interval: 1,371.38 – 1,381.14

45638 votes

Category
exclude_ties
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,407.45 rating

Confidence interval: 1,399.16 – 1,415.74

5847 votes

Category
expert
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,413.67 rating

Confidence interval: 1,399.61 – 1,427.72

2254 votes

Category
french
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,390.78 rating

Confidence interval: 1,372.07 – 1,409.50

1072 votes

Category
german
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,403.43 rating

Confidence interval: 1,399.06 – 1,407.80

40121 votes

Category
hard_prompts
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,408.60 rating

Confidence interval: 1,403.00 – 1,414.21

18214 votes

Category
hard_prompts_english
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,394.44 rating

Confidence interval: 1,388.16 – 1,400.72

12194 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,347.39 rating

Confidence interval: 1,341.07 – 1,353.70

12636 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,397.39 rating

Confidence interval: 1,388.34 – 1,406.44

4801 votes

Category
industry_legal_and_government
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,415.21 rating

Confidence interval: 1,408.55 – 1,421.87

9836 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,412.99 rating

Confidence interval: 1,402.22 – 1,423.75

3357 votes

Category
industry_mathematical
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,407.15 rating

Confidence interval: 1,397.65 – 1,416.64

4378 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,416.18 rating

Confidence interval: 1,411.18 – 1,421.19

25049 votes

Category
industry_software_and_it_services
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,369.27 rating

Confidence interval: 1,363.40 – 1,375.14

14623 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,374.90 rating

Confidence interval: 1,369.60 – 1,380.20

20269 votes

Category
instruction_following
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,368.40 rating

Confidence interval: 1,344.62 – 1,392.19

666 votes

Category
japanese
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,343.26 rating

Confidence interval: 1,324.20 – 1,362.32

1075 votes

Category
korean
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,391.71 rating

Confidence interval: 1,386.54 – 1,396.88

25673 votes

Category
longer_query
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,407.58 rating

Confidence interval: 1,397.05 – 1,418.11

3388 votes

Category
math
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,392.61 rating

Confidence interval: 1,386.03 – 1,399.18

10990 votes

Category
multi_turn
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,384.94 rating

Confidence interval: 1,380.55 – 1,389.32

35143 votes

Category
non_english
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,397.09 rating

Confidence interval: 1,393.45 – 1,400.74

61776 votes

Category
overall
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,375.99 rating

Confidence interval: 1,359.62 – 1,392.36

1288 votes

Category
polish
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,379.85 rating

Confidence interval: 1,372.31 – 1,387.39

7213 votes

Category
russian
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Qwen3.5 Flash

1,399.19 rating

Confidence interval: 1,385.09 – 1,413.29

2054 votes

Category
spanish
Evaluated configuration
Qwen3.5-Flash
Evaluation type
human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,445.60 rating

Confidence interval: 1,435.58 – 1,455.62

3673 votes

Category
chinese
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,411.83 rating

Confidence interval: 1,406.19 – 1,417.47

17719 votes

Category
coding
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,343.15 rating

Confidence interval: 1,336.23 – 1,350.07

9752 votes

Category
creative_writing
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,404.47 rating

Confidence interval: 1,399.57 – 1,409.37

26631 votes

Category
english
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,376.20 rating

Confidence interval: 1,371.31 – 1,381.08

45675 votes

Category
exclude_ties
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,407.16 rating

Confidence interval: 1,398.84 – 1,415.47

5839 votes

Category
expert
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,412.41 rating

Confidence interval: 1,398.32 – 1,426.50

2267 votes

Category
french
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,390.72 rating

Confidence interval: 1,371.97 – 1,409.47

1075 votes

Category
german
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,403.44 rating

Confidence interval: 1,399.06 – 1,407.82

40131 votes

Category
hard_prompts
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,408.71 rating

Confidence interval: 1,403.10 – 1,414.32

18209 votes

Category
hard_prompts_english
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,394.52 rating

Confidence interval: 1,388.23 – 1,400.80

12205 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,347.01 rating

Confidence interval: 1,340.68 – 1,353.33

12642 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,397.44 rating

Confidence interval: 1,388.38 – 1,406.49

4813 votes

Category
industry_legal_and_government
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,415.51 rating

Confidence interval: 1,408.85 – 1,422.18

9839 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,412.61 rating

Confidence interval: 1,401.82 – 1,423.40

3358 votes

Category
industry_mathematical
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,407.51 rating

Confidence interval: 1,398.01 – 1,417.00

4388 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,416.34 rating

Confidence interval: 1,411.33 – 1,421.35

25042 votes

Category
industry_software_and_it_services
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,368.84 rating

Confidence interval: 1,362.97 – 1,374.72

14664 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,374.54 rating

Confidence interval: 1,369.24 – 1,379.85

20300 votes

Category
instruction_following
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,368.40 rating

Confidence interval: 1,344.50 – 1,392.31

665 votes

Category
japanese
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,343.76 rating

Confidence interval: 1,324.66 – 1,362.86

1075 votes

Category
korean
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,391.74 rating

Confidence interval: 1,386.57 – 1,396.92

25685 votes

Category
longer_query
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,407.51 rating

Confidence interval: 1,396.97 – 1,418.06

3386 votes

Category
math
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,392.43 rating

Confidence interval: 1,385.85 – 1,399.02

10982 votes

Category
multi_turn
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,384.85 rating

Confidence interval: 1,380.46 – 1,389.24

35171 votes

Category
non_english
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,397.01 rating

Confidence interval: 1,393.37 – 1,400.66

61802 votes

Category
overall
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,376.24 rating

Confidence interval: 1,359.84 – 1,392.64

1286 votes

Category
polish
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,379.61 rating

Confidence interval: 1,372.06 – 1,387.16

7223 votes

Category
russian
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

Qwen3.5 Flash

1,399.74 rating

Confidence interval: 1,385.60 – 1,413.87

2052 votes

Category
spanish
Evaluated configuration
qwen3.5-flash
Evaluation type
human preference

Run it locally

Inference · memory estimate

Qwen3.5 Flash

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

API cost

USD / month

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

Qwen3.5 Flash

Capabilities

Context window
1,000,000 tokens
Maximum output
65,536 tokens
Reasoning
Yes
Tool calling
Yes
Open weights
No
Model license
Proprietary

Input modalities

  • text
  • image
  • video

Output modalities

  • text