GeekRadar / The comparison desk
COMPARE.
AI models
google

gemma-2-2b-it

AI model Released

Context window
8,192 tokens
Maximum output
8,192 tokens
Open weights
Yes

Capability & preference

Human preference · Arena

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

gemma-2-2b-it 1155.8 Gemma 2 2B IT · 95% interval 1151.7–1159.8 · 46,616 votes · 2026-10-28

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 · 2026-10-28 · rating

gemma-2-2b-it

1,132.05 rating

Confidence interval: 1,121.16 – 1,142.93

3790 votes

Category
chinese
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,111.98 rating

Confidence interval: 1,104.09 – 1,119.88

7298 votes

Category
coding
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,146.46 rating

Confidence interval: 1,138.30 – 1,154.61

7094 votes

Category
creative_writing
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,181.46 rating

Confidence interval: 1,176.46 – 1,186.46

25669 votes

Category
english
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,004.21 rating

Confidence interval: 997.56 – 1,010.87

30730 votes

Category
exclude_ties
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,095.93 rating

Confidence interval: 1,083.36 – 1,108.50

2525 votes

Category
expert
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,157.47 rating

Confidence interval: 1,127.23 – 1,187.71

469 votes

Category
french
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,114.14 rating

Confidence interval: 1,095.21 – 1,133.06

1158 votes

Category
german
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,112.40 rating

Confidence interval: 1,106.08 – 1,118.72

12322 votes

Category
hard_prompts
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,127.21 rating

Confidence interval: 1,119.49 – 1,134.94

7586 votes

Category
hard_prompts_english
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,125.62 rating

Confidence interval: 1,116.34 – 1,134.91

5197 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,126.82 rating

Confidence interval: 1,119.11 – 1,134.54

7532 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,169.58 rating

Confidence interval: 1,156.83 – 1,182.34

2655 votes

Category
industry_legal_and_government
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,158.91 rating

Confidence interval: 1,151.00 – 1,166.81

7909 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,145.58 rating

Confidence interval: 1,137.02 – 1,154.13

5632 votes

Category
industry_mathematical
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,143.59 rating

Confidence interval: 1,129.86 – 1,157.32

2336 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,132.04 rating

Confidence interval: 1,125.39 – 1,138.68

11839 votes

Category
industry_software_and_it_services
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,157.05 rating

Confidence interval: 1,150.58 – 1,163.51

12543 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,117.48 rating

Confidence interval: 1,111.93 – 1,123.03

18240 votes

Category
instruction_following
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,083.22 rating

Confidence interval: 1,064.56 – 1,101.88

1228 votes

Category
japanese
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,054.77 rating

Confidence interval: 1,029.39 – 1,080.15

717 votes

Category
korean
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,129.69 rating

Confidence interval: 1,121.16 – 1,138.22

5866 votes

Category
longer_query
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,134.52 rating

Confidence interval: 1,126.48 – 1,142.56

6599 votes

Category
math
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,117.95 rating

Confidence interval: 1,109.84 – 1,126.06

7933 votes

Category
multi_turn
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,120.59 rating

Confidence interval: 1,115.26 – 1,125.93

20947 votes

Category
non_english
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,155.75 rating

Confidence interval: 1,151.71 – 1,159.80

46616 votes

Category
overall
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,117.97 rating

Confidence interval: 1,109.19 – 1,126.74

5974 votes

Category
russian
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

gemma-2-2b-it

1,139.99 rating

Confidence interval: 1,112.52 – 1,167.46

544 votes

Category
spanish
Evaluated configuration
Gemma 2 2B IT
Evaluation type
human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating

gemma-2-2b-it

1,132.09 rating

Confidence interval: 1,121.21 – 1,142.98

3790 votes

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

gemma-2-2b-it

1,112.09 rating

Confidence interval: 1,104.19 – 1,119.98

7298 votes

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

gemma-2-2b-it

1,146.61 rating

Confidence interval: 1,138.46 – 1,154.77

7094 votes

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

gemma-2-2b-it

1,181.41 rating

Confidence interval: 1,176.41 – 1,186.41

25669 votes

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

gemma-2-2b-it

1,004.22 rating

Confidence interval: 997.57 – 1,010.88

30730 votes

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

gemma-2-2b-it

1,095.97 rating

Confidence interval: 1,083.40 – 1,108.54

2525 votes

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

gemma-2-2b-it

1,157.15 rating

Confidence interval: 1,126.91 – 1,187.40

469 votes

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

gemma-2-2b-it

1,114.13 rating

Confidence interval: 1,095.26 – 1,133.01

1158 votes

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

gemma-2-2b-it

1,112.39 rating

Confidence interval: 1,106.07 – 1,118.72

12322 votes

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

gemma-2-2b-it

1,127.10 rating

Confidence interval: 1,119.37 – 1,134.83

7586 votes

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

gemma-2-2b-it

1,125.66 rating

Confidence interval: 1,116.37 – 1,134.95

5197 votes

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

gemma-2-2b-it

1,126.89 rating

Confidence interval: 1,119.17 – 1,134.60

7532 votes

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

gemma-2-2b-it

1,169.33 rating

Confidence interval: 1,156.57 – 1,182.08

2655 votes

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

gemma-2-2b-it

1,158.89 rating

Confidence interval: 1,150.98 – 1,166.79

7909 votes

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

gemma-2-2b-it

1,145.57 rating

Confidence interval: 1,137.02 – 1,154.12

5632 votes

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

gemma-2-2b-it

1,143.65 rating

Confidence interval: 1,129.92 – 1,157.38

2336 votes

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

gemma-2-2b-it

1,132.12 rating

Confidence interval: 1,125.48 – 1,138.77

11839 votes

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

gemma-2-2b-it

1,157.11 rating

Confidence interval: 1,150.65 – 1,163.58

12543 votes

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

gemma-2-2b-it

1,117.49 rating

Confidence interval: 1,111.94 – 1,123.04

18240 votes

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

gemma-2-2b-it

1,083.06 rating

Confidence interval: 1,064.47 – 1,101.64

1228 votes

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

gemma-2-2b-it

1,054.81 rating

Confidence interval: 1,029.45 – 1,080.18

717 votes

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

gemma-2-2b-it

1,129.73 rating

Confidence interval: 1,121.21 – 1,138.26

5866 votes

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

gemma-2-2b-it

1,134.48 rating

Confidence interval: 1,126.44 – 1,142.51

6599 votes

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

gemma-2-2b-it

1,117.95 rating

Confidence interval: 1,109.84 – 1,126.06

7933 votes

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

gemma-2-2b-it

1,120.69 rating

Confidence interval: 1,115.35 – 1,126.02

20947 votes

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

gemma-2-2b-it

1,155.74 rating

Confidence interval: 1,151.69 – 1,159.78

46616 votes

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

gemma-2-2b-it

1,118.12 rating

Confidence interval: 1,109.34 – 1,126.90

5974 votes

Category
russian
Evaluated configuration
gemma-2-2b-it
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

gemma-2-2b-it

1,139.91 rating

Confidence interval: 1,112.42 – 1,167.40

544 votes

Category
spanish
Evaluated configuration
gemma-2-2b-it
Evaluation type
human preference

Run it locally

Inference · memory estimate

gemma-2-2b-it

4.9 GiB checkpoint tensor payload · 2.61 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.

Capabilities

Context window
8,192 tokens
Maximum output
8,192 tokens
Reasoning
No
Tool calling
No
Open weights
Yes
Model license
gemma
Checkpoint tensor payload
4.87 GiB

Input modalities

  • text

Output modalities

  • text