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
solar-10.7b-instruct-v1.0
1,066.60 rating
Confidence interval: 1,040.11 – 1,093.09
482 votes
Category
coding
Evaluated configuration
solar-10.7b-instruct-v1.0
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
solar-10.7b-instruct-v1.0
1,089.22 rating
Confidence interval: 1,065.24 – 1,113.19
698 votes
Category
creative_writing
Evaluated configuration
solar-10.7b-instruct-v1.0
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
solar-10.7b-instruct-v1.0
1,117.19 rating
Confidence interval: 1,102.83 – 1,131.55
3277 votes
Category
english
Evaluated configuration
solar-10.7b-instruct-v1.0
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
solar-10.7b-instruct-v1.0
907.83 rating
Confidence interval: 889.04 – 926.62
2831 votes
Category
exclude_ties
Evaluated configuration
solar-10.7b-instruct-v1.0
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
solar-10.7b-instruct-v1.0
1,061.91 rating
Confidence interval: 1,042.75 – 1,081.07
932 votes
Category
hard_prompts
Evaluated configuration
solar-10.7b-instruct-v1.0
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
solar-10.7b-instruct-v1.0
1,075.03 rating
Confidence interval: 1,053.78 – 1,096.28
765 votes
Category
hard_prompts_english
Evaluated configuration
solar-10.7b-instruct-v1.0
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating