GeekRadar / The comparison desk
COMPARE.
AI models
xai

grok-2-2024-08-13

Model evaluation

Capability & preference

Human preference · Arena

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

grok-2-2024-08-13 1304.5 grok-2-2024-08-13 · 95% interval 1300.9–1308.2 · 63,498 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-2-2024-08-13

1,289.32 rating

Confidence interval: 1,279.81 – 1,298.84

5317 votes

Category
chinese
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,287.26 rating

Confidence interval: 1,280.38 – 1,294.15

10368 votes

Category
coding
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,284.14 rating

Confidence interval: 1,276.74 – 1,291.53

9339 votes

Category
creative_writing
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,320.63 rating

Confidence interval: 1,316.08 – 1,325.18

34362 votes

Category
english
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,243.87 rating

Confidence interval: 1,238.36 – 1,249.38

40763 votes

Category
exclude_ties
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,253.54 rating

Confidence interval: 1,242.66 – 1,264.41

3541 votes

Category
expert
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,317.71 rating

Confidence interval: 1,291.40 – 1,344.02

604 votes

Category
french
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,286.84 rating

Confidence interval: 1,270.87 – 1,302.80

1609 votes

Category
german
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,271.91 rating

Confidence interval: 1,266.37 – 1,277.45

17283 votes

Category
hard_prompts
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,279.51 rating

Confidence interval: 1,272.76 – 1,286.27

10486 votes

Category
hard_prompts_english
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,265.10 rating

Confidence interval: 1,257.03 – 1,273.18

7490 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,282.06 rating

Confidence interval: 1,275.10 – 1,289.03

10293 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,325.53 rating

Confidence interval: 1,314.47 – 1,336.60

3813 votes

Category
industry_legal_and_government
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,314.36 rating

Confidence interval: 1,307.27 – 1,321.45

10796 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,283.69 rating

Confidence interval: 1,276.01 – 1,291.38

7604 votes

Category
industry_mathematical
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,306.58 rating

Confidence interval: 1,294.66 – 1,318.49

3206 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,294.41 rating

Confidence interval: 1,288.58 – 1,300.25

16648 votes

Category
industry_software_and_it_services
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,291.89 rating

Confidence interval: 1,285.98 – 1,297.80

17406 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,269.74 rating

Confidence interval: 1,264.76 – 1,274.73

25659 votes

Category
instruction_following
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,244.10 rating

Confidence interval: 1,227.31 – 1,260.90

1478 votes

Category
japanese
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,236.86 rating

Confidence interval: 1,214.88 – 1,258.84

952 votes

Category
korean
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,276.01 rating

Confidence interval: 1,268.79 – 1,283.24

8902 votes

Category
longer_query
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,283.11 rating

Confidence interval: 1,275.82 – 1,290.40

8950 votes

Category
math
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,290.16 rating

Confidence interval: 1,283.07 – 1,297.24

10743 votes

Category
multi_turn
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,282.00 rating

Confidence interval: 1,277.24 – 1,286.77

29136 votes

Category
non_english
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,304.54 rating

Confidence interval: 1,300.88 – 1,308.20

63498 votes

Category
overall
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,285.77 rating

Confidence interval: 1,278.11 – 1,293.42

8745 votes

Category
russian
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

grok-2-2024-08-13

1,281.11 rating

Confidence interval: 1,256.63 – 1,305.60

671 votes

Category
spanish
Evaluated configuration
grok-2-2024-08-13
Evaluation type
human preference

Run it locally

Inference · memory estimate

grok-2-2024-08-13

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary