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
zephyr-7b-alpha
1,052.56 rating
Confidence interval: 1,013.56 – 1,091.57
201 votes
Category
coding
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,068.60 rating
Confidence interval: 1,038.67 – 1,098.53
399 votes
Category
creative_writing
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,091.58 rating
Confidence interval: 1,074.22 – 1,108.95
1415 votes
Category
english
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating
zephyr-7b-alpha
875.23 rating
Confidence interval: 852.83 – 897.64
1218 votes
Category
exclude_ties
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,023.70 rating
Confidence interval: 994.97 – 1,052.42
333 votes
Category
hard_prompts
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,041.81 rating
Confidence interval: 1,009.42 – 1,074.21
278 votes
Category
hard_prompts_english
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,048.24 rating
Confidence interval: 1,016.50 – 1,079.97
329 votes
Category
industry_entertainment_and_sports_and_media
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,027.95 rating
Confidence interval: 993.96 – 1,061.93
294 votes
Category
industry_life_and_physical_and_social_science
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,053.25 rating
Confidence interval: 1,024.92 – 1,081.58
406 votes
Category
industry_software_and_it_services
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,081.61 rating
Confidence interval: 1,056.37 – 1,106.84
569 votes
Category
industry_writing_and_literature_and_language
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,034.23 rating
Confidence interval: 1,009.96 – 1,058.50
534 votes
Category
instruction_following
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,017.59 rating
Confidence interval: 983.13 – 1,052.05
283 votes
Category
multi_turn
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
zephyr-7b-alpha
987.82 rating
Confidence interval: 959.40 – 1,016.24
370 votes
Category
non_english
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
zephyr-7b-alpha
1,058.33 rating
Confidence interval: 1,042.67 – 1,073.98
1785 votes
Category
overall
Evaluated configuration
zephyr-7b-alpha
Evaluation type
human preference
Run it locally
Inference · memory estimate
zephyr-7b-alpha
Not enough verified architecture information for a reliable VRAM estimate.