GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

gemma-2-9b-it-simpo

Model evaluation

Capability & preference

Human preference · Arena

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

gemma-2-9b-it-simpo 1227.0 gemma-2-9b-it-simpo · 95% interval 1220.2–1233.9 · 10,072 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

gemma-2-9b-it-simpo

1,225.10 rating

Confidence interval: 1,204.25 – 1,245.95

753 votes

Category
chinese
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,190.70 rating

Confidence interval: 1,175.94 – 1,205.46

1471 votes

Category
coding
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,238.64 rating

Confidence interval: 1,224.11 – 1,253.17

1671 votes

Category
creative_writing
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,242.33 rating

Confidence interval: 1,233.66 – 1,251.00

5834 votes

Category
english
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,122.33 rating

Confidence interval: 1,111.65 – 1,133.00

6321 votes

Category
exclude_ties
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,154.31 rating

Confidence interval: 1,124.95 – 1,183.67

369 votes

Category
expert
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,218.43 rating

Confidence interval: 1,187.29 – 1,249.58

318 votes

Category
german
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,195.98 rating

Confidence interval: 1,184.68 – 1,207.28

2554 votes

Category
hard_prompts
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,200.95 rating

Confidence interval: 1,187.22 – 1,214.68

1661 votes

Category
hard_prompts_english
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,222.27 rating

Confidence interval: 1,205.12 – 1,239.43

1121 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,214.67 rating

Confidence interval: 1,200.64 – 1,228.70

1692 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,268.23 rating

Confidence interval: 1,243.80 – 1,292.67

543 votes

Category
industry_legal_and_government
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,230.85 rating

Confidence interval: 1,216.06 – 1,245.65

1596 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,195.89 rating

Confidence interval: 1,180.15 – 1,211.63

1139 votes

Category
industry_mathematical
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,221.59 rating

Confidence interval: 1,195.63 – 1,247.55

516 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,211.41 rating

Confidence interval: 1,199.54 – 1,223.28

2537 votes

Category
industry_software_and_it_services
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,225.89 rating

Confidence interval: 1,214.26 – 1,237.51

2801 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,190.91 rating

Confidence interval: 1,180.97 – 1,200.85

3741 votes

Category
instruction_following
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,139.62 rating

Confidence interval: 1,107.25 – 1,171.98

291 votes

Category
japanese
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,225.97 rating

Confidence interval: 1,207.09 – 1,244.86

841 votes

Category
longer_query
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,172.67 rating

Confidence interval: 1,157.57 – 1,187.77

1285 votes

Category
math
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,219.66 rating

Confidence interval: 1,205.21 – 1,234.11

1660 votes

Category
multi_turn
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,206.29 rating

Confidence interval: 1,196.67 – 1,215.90

4238 votes

Category
non_english
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,227.03 rating

Confidence interval: 1,220.19 – 1,233.87

10072 votes

Category
overall
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

gemma-2-9b-it-simpo

1,224.52 rating

Confidence interval: 1,207.65 – 1,241.39

1119 votes

Category
russian
Evaluated configuration
gemma-2-9b-it-simpo
Evaluation type
human preference

Run it locally

Inference · memory estimate

gemma-2-9b-it-simpo

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
MIT