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-turbos-20250226
1,323.45 rating
Confidence interval: 1,284.40 – 1,362.50
203 votes
Category
chinese
Evaluated configuration
hunyuan-turbos-20250226
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
hunyuan-turbos-20250226
1,342.41 rating
Confidence interval: 1,311.96 – 1,372.85
275 votes
Category
coding
Evaluated configuration
hunyuan-turbos-20250226
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
hunyuan-turbos-20250226
1,270.30 rating
Confidence interval: 1,244.55 – 1,296.05
424 votes
Category
creative_writing
Evaluated configuration
hunyuan-turbos-20250226
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
hunyuan-turbos-20250226
1,333.90 rating
Confidence interval: 1,319.04 – 1,348.77
1310 votes
Category
english
Evaluated configuration
hunyuan-turbos-20250226
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
hunyuan-turbos-20250226
1,265.19 rating
Confidence interval: 1,246.38 – 1,283.99
1373 votes
Category
exclude_ties
Evaluated configuration
hunyuan-turbos-20250226
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
hunyuan-turbos-20250226
1,324.71 rating
Confidence interval: 1,302.20 – 1,347.21
531 votes
Category
hard_prompts
Evaluated configuration
hunyuan-turbos-20250226
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
hunyuan-turbos-20250226
1,337.08 rating
Confidence interval: 1,309.32 – 1,364.83
367 votes
Category
hard_prompts_english
Evaluated configuration
hunyuan-turbos-20250226
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating