GeekRadar / The comparison desk
COMPARE.
AI models
google

Gemini 2.5 Flash Preview (09-2025)

Model evaluation

Capability & preference

Human preference · Arena

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

Gemini 2.5 Flash Preview (09-2025) 1406.5 Gemini 2.5 Flash Preview (09-2025) · 95% interval 1402.5–1410.5 · 32,996 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

Gemini 2.5 Flash Preview (09-2025)

1,449.13 rating

Confidence interval: 1,437.01 – 1,461.25

2375 votes

Category
chinese
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,400.27 rating

Confidence interval: 1,392.86 – 1,407.68

6911 votes

Category
coding
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,386.34 rating

Confidence interval: 1,377.33 – 1,395.35

4552 votes

Category
creative_writing
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,412.82 rating

Confidence interval: 1,407.25 – 1,418.39

14039 votes

Category
english
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,388.87 rating

Confidence interval: 1,383.34 – 1,394.39

23426 votes

Category
exclude_ties
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,418.06 rating

Confidence interval: 1,404.20 – 1,431.91

1788 votes

Category
expert
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,433.23 rating

Confidence interval: 1,408.58 – 1,457.88

588 votes

Category
french
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,418.03 rating

Confidence interval: 1,394.87 – 1,441.20

657 votes

Category
german
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,408.83 rating

Confidence interval: 1,403.70 – 1,413.96

17741 votes

Category
hard_prompts
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,410.35 rating

Confidence interval: 1,403.20 – 1,417.49

7542 votes

Category
hard_prompts_english
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,397.94 rating

Confidence interval: 1,390.11 – 1,405.76

6149 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,385.69 rating

Confidence interval: 1,377.84 – 1,393.54

6071 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,427.52 rating

Confidence interval: 1,414.98 – 1,440.05

2154 votes

Category
industry_legal_and_government
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,434.31 rating

Confidence interval: 1,426.00 – 1,442.63

5221 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,419.80 rating

Confidence interval: 1,405.78 – 1,433.83

1704 votes

Category
industry_mathematical
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,424.86 rating

Confidence interval: 1,410.44 – 1,439.28

1703 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,408.46 rating

Confidence interval: 1,402.46 – 1,414.47

11704 votes

Category
industry_software_and_it_services
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,395.01 rating

Confidence interval: 1,387.73 – 1,402.30

7380 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,393.67 rating

Confidence interval: 1,387.12 – 1,400.21

9353 votes

Category
instruction_following
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,404.92 rating

Confidence interval: 1,373.46 – 1,436.38

359 votes

Category
japanese
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,384.47 rating

Confidence interval: 1,361.76 – 1,407.19

703 votes

Category
korean
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,405.38 rating

Confidence interval: 1,398.48 – 1,412.28

8234 votes

Category
longer_query
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,415.40 rating

Confidence interval: 1,402.48 – 1,428.33

1977 votes

Category
math
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,396.33 rating

Confidence interval: 1,388.28 – 1,404.39

5748 votes

Category
multi_turn
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,395.72 rating

Confidence interval: 1,390.72 – 1,400.72

18908 votes

Category
non_english
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,406.47 rating

Confidence interval: 1,402.45 – 1,410.49

32996 votes

Category
overall
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,409.53 rating

Confidence interval: 1,393.11 – 1,425.96

1274 votes

Category
polish
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,392.12 rating

Confidence interval: 1,381.64 – 1,402.60

2945 votes

Category
russian
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Gemini 2.5 Flash Preview (09-2025)

1,404.75 rating

Confidence interval: 1,385.13 – 1,424.37

980 votes

Category
spanish
Evaluated configuration
Gemini 2.5 Flash Preview (09-2025)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Gemini 2.5 Flash Preview (09-2025)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary