GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

yi-lightning

Model evaluation

Capability & preference

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

yi-lightning 1301.5 yi-lightning · 95% interval 1296.6–1306.4 · 27,332 votes · 2026-10-15

Recorded results

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

yi-lightning

1,319.85 rating

Confidence interval: 1,307.50 – 1,332.20

2736 votes

Category
chinese
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

yi-lightning

1,312.18 rating

Confidence interval: 1,302.57 – 1,321.79

4316 votes

Category
coding
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

yi-lightning

1,279.94 rating

Confidence interval: 1,269.55 – 1,290.32

3781 votes

Category
creative_writing
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

yi-lightning

1,329.69 rating

Confidence interval: 1,323.16 – 1,336.21

13650 votes

Category
english
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

yi-lightning

1,241.15 rating

Confidence interval: 1,233.61 – 1,248.69

17462 votes

Category
exclude_ties
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

yi-lightning

1,285.33 rating

Confidence interval: 1,270.74 – 1,299.92

1533 votes

Category
expert
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

yi-lightning

1,305.32 rating

Confidence interval: 1,268.43 – 1,342.22

270 votes

Category
french
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

yi-lightning

1,266.74 rating

Confidence interval: 1,244.87 – 1,288.61

700 votes

Category
german
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

yi-lightning

1,302.00 rating

Confidence interval: 1,294.08 – 1,309.92

6961 votes

Category
hard_prompts
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

yi-lightning

1,321.92 rating

Confidence interval: 1,312.13 – 1,331.71

3939 votes

Category
hard_prompts_english
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

yi-lightning

1,271.13 rating

Confidence interval: 1,260.10 – 1,282.16

3420 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

yi-lightning

1,272.19 rating

Confidence interval: 1,262.09 – 1,282.29

3980 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

yi-lightning

1,327.61 rating

Confidence interval: 1,312.73 – 1,342.49

1697 votes

Category
industry_legal_and_government
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

yi-lightning

1,315.44 rating

Confidence interval: 1,305.82 – 1,325.06

4822 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

yi-lightning

1,306.37 rating

Confidence interval: 1,295.83 – 1,316.91

3206 votes

Category
industry_mathematical
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

yi-lightning

1,306.06 rating

Confidence interval: 1,289.93 – 1,322.19

1451 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

yi-lightning

1,313.88 rating

Confidence interval: 1,305.70 – 1,322.06

7070 votes

Category
industry_software_and_it_services
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

yi-lightning

1,276.30 rating

Confidence interval: 1,268.11 – 1,284.49

7632 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

yi-lightning

1,278.16 rating

Confidence interval: 1,271.26 – 1,285.05

10932 votes

Category
instruction_following
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

yi-lightning

1,227.98 rating

Confidence interval: 1,203.67 – 1,252.28

565 votes

Category
japanese
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

yi-lightning

1,192.42 rating

Confidence interval: 1,161.35 – 1,223.49

425 votes

Category
korean
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

yi-lightning

1,296.62 rating

Confidence interval: 1,286.13 – 1,307.12

3634 votes

Category
longer_query
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

yi-lightning

1,299.60 rating

Confidence interval: 1,289.78 – 1,309.41

3921 votes

Category
math
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

yi-lightning

1,311.09 rating

Confidence interval: 1,301.21 – 1,320.97

4748 votes

Category
multi_turn
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

yi-lightning

1,269.01 rating

Confidence interval: 1,262.57 – 1,275.45

13682 votes

Category
non_english
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

yi-lightning

1,301.49 rating

Confidence interval: 1,296.60 – 1,306.39

27332 votes

Category
overall
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

yi-lightning

1,255.08 rating

Confidence interval: 1,245.12 – 1,265.05

4189 votes

Category
russian
Evaluated configuration
yi-lightning
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

yi-lightning

1,315.33 rating

Confidence interval: 1,283.45 – 1,347.22

341 votes

Category
spanish
Evaluated configuration
yi-lightning
Evaluation type
human preference

Run it locally

Inference · memory estimate

yi-lightning

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary