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
Ministral 8B (2410)
1,193.31 rating
Confidence interval: 1,168.18 – 1,218.45
478 votes
Category
chinese
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Ministral 8B (2410)
1,201.61 rating
Confidence interval: 1,183.07 – 1,220.15
838 votes
Category
coding
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Ministral 8B (2410)
1,174.46 rating
Confidence interval: 1,153.02 – 1,195.89
678 votes
Category
creative_writing
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Ministral 8B (2410)
1,212.42 rating
Confidence interval: 1,200.53 – 1,224.32
2322 votes
Category
english
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Ministral 8B (2410)
1,060.31 rating
Confidence interval: 1,045.50 – 1,075.12
3020 votes
Category
exclude_ties
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Ministral 8B (2410)
1,169.57 rating
Confidence interval: 1,139.89 – 1,199.24
332 votes
Category
expert
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Ministral 8B (2410)
1,191.07 rating
Confidence interval: 1,176.02 – 1,206.11
1257 votes
Category
hard_prompts
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Ministral 8B (2410)
1,206.11 rating
Confidence interval: 1,186.09 – 1,226.13
703 votes
Category
hard_prompts_english
Evaluated configuration
Ministral 8B (2410)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating