GeekRadar / The comparison desk
COMPARE.
AI models
tencent

hunyuan-hy3-preview

Model evaluation

Capability & preference

Human preference · Arena

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

hunyuan-hy3-preview 1401.5 hunyuan-hy3-preview · 95% interval 1394.2–1408.9 · 6,879 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

hunyuan-hy3-preview

1,433.94 rating

Confidence interval: 1,405.25 – 1,462.62

454 votes

Category
chinese
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,421.48 rating

Confidence interval: 1,408.08 – 1,434.88

2055 votes

Category
coding
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,339.67 rating

Confidence interval: 1,320.75 – 1,358.59

1046 votes

Category
creative_writing
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,409.70 rating

Confidence interval: 1,399.40 – 1,420.00

3339 votes

Category
english
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,383.82 rating

Confidence interval: 1,374.00 – 1,393.63

5189 votes

Category
exclude_ties
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,421.91 rating

Confidence interval: 1,400.07 – 1,443.75

710 votes

Category
expert
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,426.08 rating

Confidence interval: 1,386.71 – 1,465.45

237 votes

Category
french
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,413.20 rating

Confidence interval: 1,404.15 – 1,422.25

4589 votes

Category
hard_prompts
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,416.17 rating

Confidence interval: 1,403.68 – 1,428.67

2317 votes

Category
hard_prompts_english
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,397.54 rating

Confidence interval: 1,381.73 – 1,413.34

1433 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,352.59 rating

Confidence interval: 1,335.98 – 1,369.20

1373 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,429.88 rating

Confidence interval: 1,404.23 – 1,455.52

554 votes

Category
industry_legal_and_government
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,415.49 rating

Confidence interval: 1,398.48 – 1,432.50

1239 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,443.19 rating

Confidence interval: 1,415.86 – 1,470.51

451 votes

Category
industry_mathematical
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,405.16 rating

Confidence interval: 1,377.52 – 1,432.80

481 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,423.38 rating

Confidence interval: 1,411.98 – 1,434.78

2802 votes

Category
industry_software_and_it_services
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,370.56 rating

Confidence interval: 1,355.39 – 1,385.72

1602 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,383.59 rating

Confidence interval: 1,371.08 – 1,396.10

2339 votes

Category
instruction_following
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,401.99 rating

Confidence interval: 1,390.68 – 1,413.30

3054 votes

Category
longer_query
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,424.31 rating

Confidence interval: 1,397.64 – 1,450.97

431 votes

Category
math
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,410.77 rating

Confidence interval: 1,393.21 – 1,428.33

1207 votes

Category
multi_turn
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,390.36 rating

Confidence interval: 1,380.22 – 1,400.49

3540 votes

Category
non_english
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,401.54 rating

Confidence interval: 1,394.15 – 1,408.92

6879 votes

Category
overall
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,398.10 rating

Confidence interval: 1,375.43 – 1,420.78

708 votes

Category
russian
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

hunyuan-hy3-preview

1,398.74 rating

Confidence interval: 1,358.84 – 1,438.63

210 votes

Category
spanish
Evaluated configuration
hunyuan-hy3-preview
Evaluation type
human preference

Run it locally

Inference · memory estimate

hunyuan-hy3-preview

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
tencent-hunyuan-community