GeekRadar / The comparison desk
COMPARE.
AI models
google

gemini-3-flash (thinking-minimal)

Model evaluation

Capability & preference

Human preference · Arena

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

gemini-3-flash (thinking-minimal) 1442.4 gemini-3-flash (thinking-minimal) · 95% interval 1439.4–1445.4 · 91,039 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

gemini-3-flash (thinking-minimal)

1,476.91 rating

Confidence interval: 1,468.75 – 1,485.08

5997 votes

Category
chinese
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,435.92 rating

Confidence interval: 1,430.96 – 1,440.88

23982 votes

Category
coding
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,435.36 rating

Confidence interval: 1,429.56 – 1,441.17

15198 votes

Category
creative_writing
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,440.67 rating

Confidence interval: 1,436.55 – 1,444.79

40280 votes

Category
english
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,438.91 rating

Confidence interval: 1,434.82 – 1,442.99

67574 votes

Category
exclude_ties
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,419.46 rating

Confidence interval: 1,412.39 – 1,426.53

8588 votes

Category
expert
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,445.22 rating

Confidence interval: 1,432.51 – 1,457.92

2851 votes

Category
french
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,465.67 rating

Confidence interval: 1,449.31 – 1,482.02

1474 votes

Category
german
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,437.68 rating

Confidence interval: 1,433.94 – 1,441.42

57574 votes

Category
hard_prompts
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,432.24 rating

Confidence interval: 1,427.40 – 1,437.09

26220 votes

Category
hard_prompts_english
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,422.31 rating

Confidence interval: 1,416.99 – 1,427.64

18145 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,420.76 rating

Confidence interval: 1,415.45 – 1,426.06

19464 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,445.05 rating

Confidence interval: 1,437.50 – 1,452.61

7229 votes

Category
industry_legal_and_government
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,454.39 rating

Confidence interval: 1,448.81 – 1,459.97

14947 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,431.20 rating

Confidence interval: 1,422.11 – 1,440.29

4877 votes

Category
industry_mathematical
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,446.80 rating

Confidence interval: 1,438.91 – 1,454.68

6675 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,444.66 rating

Confidence interval: 1,440.33 – 1,448.99

35022 votes

Category
industry_software_and_it_services
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,436.21 rating

Confidence interval: 1,431.17 – 1,441.26

22031 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,414.17 rating

Confidence interval: 1,409.57 – 1,418.77

29800 votes

Category
instruction_following
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,424.61 rating

Confidence interval: 1,402.62 – 1,446.59

852 votes

Category
japanese
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,417.62 rating

Confidence interval: 1,400.80 – 1,434.44

1413 votes

Category
korean
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,428.75 rating

Confidence interval: 1,424.22 – 1,433.28

36512 votes

Category
longer_query
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,447.74 rating

Confidence interval: 1,439.10 – 1,456.38

4994 votes

Category
math
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,454.71 rating

Confidence interval: 1,449.14 – 1,460.29

16296 votes

Category
multi_turn
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,437.77 rating

Confidence interval: 1,434.01 – 1,441.53

50758 votes

Category
non_english
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,442.38 rating

Confidence interval: 1,439.37 – 1,445.39

91039 votes

Category
overall
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,468.19 rating

Confidence interval: 1,454.72 – 1,481.67

1911 votes

Category
polish
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,453.73 rating

Confidence interval: 1,447.35 – 1,460.11

10627 votes

Category
russian
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

gemini-3-flash (thinking-minimal)

1,426.65 rating

Confidence interval: 1,414.39 – 1,438.92

2772 votes

Category
spanish
Evaluated configuration
gemini-3-flash (thinking-minimal)
Evaluation type
human preference

Run it locally

Inference · memory estimate

gemini-3-flash (thinking-minimal)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary