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
grok-4-0709
1,426.57 rating
Confidence interval: 1,415.23 – 1,437.91
2828 votes
Category
chinese
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating
grok-4-0709
1,408.16 rating
Confidence interval: 1,401.25 – 1,415.07
8041 votes
Category
coding
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
grok-4-0709
1,396.36 rating
Confidence interval: 1,387.95 – 1,404.77
5344 votes
Category
creative_writing
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
grok-4-0709
1,418.01 rating
Confidence interval: 1,412.83 – 1,423.19
17566 votes
Category
english
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
grok-4-0709
1,395.36 rating
Confidence interval: 1,390.06 – 1,400.67
28660 votes
Category
exclude_ties
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating
grok-4-0709
1,414.83 rating
Confidence interval: 1,401.74 – 1,427.93
2073 votes
Category
expert
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating
grok-4-0709
1,417.90 rating
Confidence interval: 1,392.97 – 1,442.83
591 votes
Category
french
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating
grok-4-0709
1,429.02 rating
Confidence interval: 1,409.00 – 1,449.03
904 votes
Category
german
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
grok-4-0709
1,408.57 rating
Confidence interval: 1,403.56 – 1,413.57
19381 votes
Category
hard_prompts
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
grok-4-0709
1,410.15 rating
Confidence interval: 1,403.45 – 1,416.86
8814 votes
Category
hard_prompts_english
Evaluated configuration
grok-4-0709
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating