GeekRadar / The comparison desk
COMPARE.
AI models
xai

grok-4.1-thinking

Model evaluation

Capability & preference

Human preference · Arena

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

grok-4.1-thinking 1436.5 grok-4.1-thinking · 95% interval 1433.4–1439.6 · 66,509 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

grok-4.1-thinking

1,451.20 rating

Confidence interval: 1,442.51 – 1,459.89

5063 votes

Category
chinese
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,445.10 rating

Confidence interval: 1,439.62 – 1,450.58

15421 votes

Category
coding
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,403.52 rating

Confidence interval: 1,397.00 – 1,410.03

10145 votes

Category
creative_writing
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,442.62 rating

Confidence interval: 1,438.27 – 1,446.96

28201 votes

Category
english
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,430.93 rating

Confidence interval: 1,426.68 – 1,435.18

49071 votes

Category
exclude_ties
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,417.50 rating

Confidence interval: 1,408.80 – 1,426.19

4933 votes

Category
expert
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,447.91 rating

Confidence interval: 1,432.38 – 1,463.44

1716 votes

Category
french
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,439.17 rating

Confidence interval: 1,421.19 – 1,457.15

1144 votes

Category
german
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,433.14 rating

Confidence interval: 1,429.20 – 1,437.09

38533 votes

Category
hard_prompts
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,435.56 rating

Confidence interval: 1,430.26 – 1,440.86

16633 votes

Category
hard_prompts_english
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,415.79 rating

Confidence interval: 1,409.89 – 1,421.69

12793 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,406.84 rating

Confidence interval: 1,400.87 – 1,412.81

12775 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,437.74 rating

Confidence interval: 1,428.96 – 1,446.52

4923 votes

Category
industry_legal_and_government
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,436.64 rating

Confidence interval: 1,430.45 – 1,442.84

11136 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,414.48 rating

Confidence interval: 1,403.87 – 1,425.09

3255 votes

Category
industry_mathematical
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,450.20 rating

Confidence interval: 1,440.81 – 1,459.59

4484 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,450.67 rating

Confidence interval: 1,445.99 – 1,455.34

23784 votes

Category
industry_software_and_it_services
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,404.33 rating

Confidence interval: 1,398.83 – 1,409.83

15309 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,394.98 rating

Confidence interval: 1,389.95 – 1,400.01

19318 votes

Category
instruction_following
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,375.91 rating

Confidence interval: 1,351.97 – 1,399.85

641 votes

Category
japanese
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,394.58 rating

Confidence interval: 1,375.92 – 1,413.25

1123 votes

Category
korean
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,406.84 rating

Confidence interval: 1,401.89 – 1,411.79

21143 votes

Category
longer_query
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,421.91 rating

Confidence interval: 1,412.30 – 1,431.51

3950 votes

Category
math
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,430.43 rating

Confidence interval: 1,424.30 – 1,436.55

12245 votes

Category
multi_turn
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,425.11 rating

Confidence interval: 1,421.18 – 1,429.04

38307 votes

Category
non_english
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,436.50 rating

Confidence interval: 1,433.36 – 1,439.64

66509 votes

Category
overall
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,454.89 rating

Confidence interval: 1,441.18 – 1,468.61

1940 votes

Category
polish
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,433.78 rating

Confidence interval: 1,426.68 – 1,440.89

7940 votes

Category
russian
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

grok-4.1-thinking

1,438.27 rating

Confidence interval: 1,424.11 – 1,452.44

1949 votes

Category
spanish
Evaluated configuration
grok-4.1-thinking
Evaluation type
human preference

Run it locally

Inference · memory estimate

grok-4.1-thinking

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary