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 / codinghuman preference · 2026-10-15 · rating
hunyuan-vision-1.5-thinking
1,419.90 rating
Confidence interval: 1,392.53 – 1,447.27
419 votes
Category
coding
Evaluated configuration
hunyuan-vision-1.5-thinking
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
hunyuan-vision-1.5-thinking
1,387.85 rating
Confidence interval: 1,351.45 – 1,424.25
250 votes
Category
creative_writing
Evaluated configuration
hunyuan-vision-1.5-thinking
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
hunyuan-vision-1.5-thinking
1,426.59 rating
Confidence interval: 1,408.21 – 1,444.97
959 votes
Category
english
Evaluated configuration
hunyuan-vision-1.5-thinking
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
hunyuan-vision-1.5-thinking
1,386.67 rating
Confidence interval: 1,369.19 – 1,404.16
1560 votes
Category
exclude_ties
Evaluated configuration
hunyuan-vision-1.5-thinking
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
hunyuan-vision-1.5-thinking
1,416.72 rating
Confidence interval: 1,399.52 – 1,433.93
1077 votes
Category
hard_prompts
Evaluated configuration
hunyuan-vision-1.5-thinking
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
hunyuan-vision-1.5-thinking
1,436.64 rating
Confidence interval: 1,411.54 – 1,461.73
491 votes
Category
hard_prompts_english
Evaluated configuration
hunyuan-vision-1.5-thinking
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating