GeekRadar / The comparison desk
COMPARE.
AI models
xai

grok-3-mini-high

Model evaluation

Capability & preference

Human preference · Arena

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

grok-3-mini-high 1369.5 grok-3-mini-high · 95% interval 1364.1–1375.0 · 15,596 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-3-mini-high

1,386.74 rating

Confidence interval: 1,366.78 – 1,406.70

870 votes

Category
chinese
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

grok-3-mini-high

1,379.15 rating

Confidence interval: 1,368.76 – 1,389.53

3163 votes

Category
coding
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

grok-3-mini-high

1,334.42 rating

Confidence interval: 1,320.82 – 1,348.03

1908 votes

Category
creative_writing
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,379.02 rating

Confidence interval: 1,371.76 – 1,386.28

7124 votes

Category
english
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

grok-3-mini-high

1,335.21 rating

Confidence interval: 1,327.85 – 1,342.57

11114 votes

Category
exclude_ties
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

grok-3-mini-high

1,394.40 rating

Confidence interval: 1,375.53 – 1,413.27

911 votes

Category
expert
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,346.84 rating

Confidence interval: 1,303.43 – 1,390.26

170 votes

Category
french
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,349.43 rating

Confidence interval: 1,320.44 – 1,378.42

416 votes

Category
german
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

grok-3-mini-high

1,374.44 rating

Confidence interval: 1,367.23 – 1,381.64

7273 votes

Category
hard_prompts
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,377.74 rating

Confidence interval: 1,367.51 – 1,387.96

3388 votes

Category
hard_prompts_english
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

grok-3-mini-high

1,343.71 rating

Confidence interval: 1,332.20 – 1,355.23

2639 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

grok-3-mini-high

1,335.67 rating

Confidence interval: 1,324.75 – 1,346.60

2850 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

grok-3-mini-high

1,391.00 rating

Confidence interval: 1,373.14 – 1,408.85

1088 votes

Category
industry_legal_and_government
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

grok-3-mini-high

1,381.56 rating

Confidence interval: 1,369.95 – 1,393.16

2582 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,368.52 rating

Confidence interval: 1,350.14 – 1,386.91

960 votes

Category
industry_mathematical
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

grok-3-mini-high

1,390.40 rating

Confidence interval: 1,371.25 – 1,409.55

954 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

grok-3-mini-high

1,378.83 rating

Confidence interval: 1,370.58 – 1,387.07

5320 votes

Category
industry_software_and_it_services
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

grok-3-mini-high

1,345.93 rating

Confidence interval: 1,335.84 – 1,356.01

3420 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

grok-3-mini-high

1,356.65 rating

Confidence interval: 1,347.49 – 1,365.82

4156 votes

Category
instruction_following
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

grok-3-mini-high

1,317.22 rating

Confidence interval: 1,287.39 – 1,347.06

398 votes

Category
japanese
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,335.30 rating

Confidence interval: 1,300.18 – 1,370.43

296 votes

Category
korean
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,371.68 rating

Confidence interval: 1,361.30 – 1,382.05

3229 votes

Category
longer_query
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,385.84 rating

Confidence interval: 1,367.61 – 1,404.06

969 votes

Category
math
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,355.16 rating

Confidence interval: 1,343.35 – 1,366.96

2489 votes

Category
multi_turn
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,352.11 rating

Confidence interval: 1,345.42 – 1,358.81

8467 votes

Category
non_english
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,369.53 rating

Confidence interval: 1,364.09 – 1,374.96

15596 votes

Category
overall
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,355.78 rating

Confidence interval: 1,340.46 – 1,371.10

1572 votes

Category
polish
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,351.09 rating

Confidence interval: 1,334.25 – 1,367.93

1146 votes

Category
russian
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

grok-3-mini-high

1,359.34 rating

Confidence interval: 1,327.28 – 1,391.41

322 votes

Category
spanish
Evaluated configuration
grok-3-mini-high
Evaluation type
human preference

Run it locally

Inference · memory estimate

grok-3-mini-high

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary