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
Phi-3-small-8k-instruct
1,060.70 rating
Confidence interval: 1,046.68 – 1,074.72
2376 votes
Category
chinese
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,101.13 rating
Confidence interval: 1,089.61 – 1,112.65
3219 votes
Category
coding
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,082.50 rating
Confidence interval: 1,069.58 – 1,095.42
2400 votes
Category
creative_writing
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,150.54 rating
Confidence interval: 1,142.68 – 1,158.40
9107 votes
Category
english
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
935.96 rating
Confidence interval: 925.67 – 946.25
11205 votes
Category
exclude_ties
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,066.88 rating
Confidence interval: 1,047.80 – 1,085.96
895 votes
Category
expert
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,135.11 rating
Confidence interval: 1,098.99 – 1,171.22
271 votes
Category
french
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,079.67 rating
Confidence interval: 1,055.20 – 1,104.14
620 votes
Category
german
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,100.18 rating
Confidence interval: 1,090.75 – 1,109.61
5229 votes
Category
hard_prompts
Evaluated configuration
Phi-3 Small 8K Instruct
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating
Phi-3-small-8k-instruct
1,125.21 rating
Confidence interval: 1,113.66 – 1,136.77
3070 votes
Category
hard_prompts_english
Evaluated configuration
Phi-3 Small 8K Instruct
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
Phi-3-small-8k-instruct
1,055.48 rating
Confidence interval: 1,043.06 – 1,067.89
2625 votes
Category
industry_entertainment_and_sports_and_media
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,102.34 rating
Confidence interval: 1,084.19 – 1,120.49
1187 votes
Category
industry_legal_and_government
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,106.98 rating
Confidence interval: 1,094.72 – 1,119.24
3117 votes
Category
industry_life_and_physical_and_social_science
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,152.95 rating
Confidence interval: 1,139.36 – 1,166.54
1849 votes
Category
industry_mathematical
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,097.12 rating
Confidence interval: 1,077.76 – 1,116.48
977 votes
Category
industry_medicine_and_healthcare
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,108.42 rating
Confidence interval: 1,098.64 – 1,118.19
5136 votes
Category
industry_software_and_it_services
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,093.71 rating
Confidence interval: 1,083.71 – 1,103.71
4639 votes
Category
industry_writing_and_literature_and_language
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,086.49 rating
Confidence interval: 1,077.88 – 1,095.10
6632 votes
Category
instruction_following
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
965.64 rating
Confidence interval: 933.99 – 997.29
390 votes
Category
japanese
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
893.85 rating
Confidence interval: 868.29 – 919.42
726 votes
Category
korean
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,087.39 rating
Confidence interval: 1,074.25 – 1,100.53
2238 votes
Category
longer_query
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,150.70 rating
Confidence interval: 1,138.01 – 1,163.39
2092 votes
Category
math
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,068.39 rating
Confidence interval: 1,056.21 – 1,080.57
3115 votes
Category
multi_turn
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,057.93 rating
Confidence interval: 1,049.82 – 1,066.04
8659 votes
Category
non_english
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,109.78 rating
Confidence interval: 1,103.94 – 1,115.63
17766 votes
Category
overall
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,110.47 rating
Confidence interval: 1,097.35 – 1,123.60
2195 votes
Category
russian
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating
Phi-3-small-8k-instruct
1,111.06 rating
Confidence interval: 1,073.65 – 1,148.47
272 votes
Category
spanish
Evaluated configuration
phi-3-small-8k-instruct
Evaluation type
human preference
Run it locally
Inference · memory estimate
Phi-3-small-8k-instruct
13.8 GiB checkpoint tensor payload · 7.39 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.