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
gemma-3n-E4B-it
1,308.80 rating
Confidence interval: 1,293.20 – 1,324.40
1489 votes
Category
chinese
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,268.11 rating
Confidence interval: 1,257.87 – 1,278.35
3488 votes
Category
coding
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,286.83 rating
Confidence interval: 1,275.71 – 1,297.94
2853 votes
Category
creative_writing
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,321.33 rating
Confidence interval: 1,314.70 – 1,327.96
10096 votes
Category
english
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,244.91 rating
Confidence interval: 1,237.83 – 1,251.99
15702 votes
Category
exclude_ties
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,245.53 rating
Confidence interval: 1,227.50 – 1,263.55
1076 votes
Category
expert
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,330.35 rating
Confidence interval: 1,297.33 – 1,363.38
327 votes
Category
french
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,311.42 rating
Confidence interval: 1,289.76 – 1,333.08
719 votes
Category
german
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,283.85 rating
Confidence interval: 1,276.77 – 1,290.93
8472 votes
Category
hard_prompts
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
gemma-3n-E4B-it
1,291.60 rating
Confidence interval: 1,282.14 – 1,301.06
4113 votes
Category
hard_prompts_english
Evaluated configuration
Gemma 3n E4B IT
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,262.35 rating
Confidence interval: 1,252.87 – 1,271.82
3872 votes
Category
industry_entertainment_and_sports_and_media
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,309.19 rating
Confidence interval: 1,293.45 – 1,324.92
1389 votes
Category
industry_legal_and_government
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,323.51 rating
Confidence interval: 1,313.53 – 1,333.50
3676 votes
Category
industry_life_and_physical_and_social_science
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,272.79 rating
Confidence interval: 1,257.08 – 1,288.51
1402 votes
Category
industry_mathematical
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,320.40 rating
Confidence interval: 1,303.89 – 1,336.92
1320 votes
Category
industry_medicine_and_healthcare
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,307.53 rating
Confidence interval: 1,299.73 – 1,315.33
6412 votes
Category
industry_software_and_it_services
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,286.86 rating
Confidence interval: 1,278.14 – 1,295.59
4861 votes
Category
industry_writing_and_literature_and_language
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,255.05 rating
Confidence interval: 1,246.47 – 1,263.62
4969 votes
Category
instruction_following
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,271.41 rating
Confidence interval: 1,248.88 – 1,293.94
734 votes
Category
japanese
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,259.22 rating
Confidence interval: 1,232.63 – 1,285.81
512 votes
Category
korean
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,275.89 rating
Confidence interval: 1,264.95 – 1,286.83
2877 votes
Category
longer_query
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,250.80 rating
Confidence interval: 1,235.91 – 1,265.68
1562 votes
Category
math
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,276.34 rating
Confidence interval: 1,265.85 – 1,286.83
3282 votes
Category
multi_turn
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,285.07 rating
Confidence interval: 1,278.85 – 1,291.28
12035 votes
Category
non_english
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,305.57 rating
Confidence interval: 1,300.42 – 1,310.73
22137 votes
Category
overall
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,285.35 rating
Confidence interval: 1,271.12 – 1,299.58
1921 votes
Category
polish
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,288.35 rating
Confidence interval: 1,274.26 – 1,302.44
1632 votes
Category
russian
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
gemma-3n-E4B-it
1,304.67 rating
Confidence interval: 1,276.07 – 1,333.27
451 votes
Category
spanish
Evaluated configuration
gemma-3n-e4b-it
Evaluation type
human preference
Run it locally
Inference · memory estimate
gemma-3n-E4B-it
14.6 GiB checkpoint tensor payload · 7.85 B stored tensor elements · BF16
Stored tensors can include buffers and repeated shared weights. A verified unique parameter count is unavailable. Checkpoint size is not the memory required to run the model.