GeekRadar / The comparison desk

COMPARE.

AI models

xai

Grok 4.20 Multi-Agent Beta (0309)

Model evaluation

Capability & preference

Human preference · Arena

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

Grok 4.20 Multi-Agent Beta (0309) 1450.7 Grok 4.20 Multi-Agent Beta (0309) · 95% interval 1447.0–1454.3 · 64,917 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

Grok 4.20 Multi-Agent Beta (0309)

1,475.94 rating

Confidence interval: 1,466.13 – 1,485.75

4036 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,457.42 rating

Confidence interval: 1,451.73 – 1,463.11

18016 votes

Oct 28, 2026

Category: coding
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,436.29 rating

Confidence interval: 1,429.55 – 1,443.02

11285 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,452.40 rating

Confidence interval: 1,447.53 – 1,457.26

29595 votes

Oct 28, 2026

Category: english
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,448.15 rating

Confidence interval: 1,443.36 – 1,452.94

49108 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / experthuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,446.03 rating

Confidence interval: 1,438.03 – 1,454.04

6549 votes

Oct 28, 2026

Category: expert
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / frenchhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,466.43 rating

Confidence interval: 1,452.40 – 1,480.45

2263 votes

Oct 28, 2026

Category: french
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / germanhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,457.31 rating

Confidence interval: 1,438.14 – 1,476.47

1058 votes

Oct 28, 2026

Category: german
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,448.28 rating

Confidence interval: 1,443.86 – 1,452.70

42159 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,446.18 rating

Confidence interval: 1,440.64 – 1,451.73

19974 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,422.69 rating

Confidence interval: 1,416.44 – 1,428.95

12893 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,422.35 rating

Confidence interval: 1,416.15 – 1,428.55

14196 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,448.31 rating

Confidence interval: 1,439.47 – 1,457.14

5184 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,455.09 rating

Confidence interval: 1,448.49 – 1,461.68

10678 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,441.62 rating

Confidence interval: 1,431.19 – 1,452.05

3558 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,452.40 rating

Confidence interval: 1,443.22 – 1,461.58

4835 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,461.13 rating

Confidence interval: 1,456.06 – 1,466.20

25579 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,430.80 rating

Confidence interval: 1,424.94 – 1,436.66

16153 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,420.24 rating

Confidence interval: 1,414.93 – 1,425.54

21950 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / japanesehuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,406.25 rating

Confidence interval: 1,381.61 – 1,430.88

626 votes

Oct 28, 2026

Category: japanese
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / koreanhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,414.81 rating

Confidence interval: 1,395.54 – 1,434.08

1047 votes

Oct 28, 2026

Category: korean
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,431.03 rating

Confidence interval: 1,425.87 – 1,436.19

27978 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,440.61 rating

Confidence interval: 1,430.15 – 1,451.06

3391 votes

Oct 28, 2026

Category: math
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,452.12 rating

Confidence interval: 1,445.47 – 1,458.77

10620 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,440.54 rating

Confidence interval: 1,436.03 – 1,445.05

35321 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,450.65 rating

Confidence interval: 1,446.97 – 1,454.33

64917 votes

Oct 28, 2026

Category: overall
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / polishhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,453.99 rating

Confidence interval: 1,437.79 – 1,470.19

1310 votes

Oct 28, 2026

Category: polish
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / russianhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,457.24 rating

Confidence interval: 1,449.54 – 1,464.93

7153 votes

Oct 28, 2026

Category: russian
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Arena text preference / spanishhuman preference · 2026-10-28 · rating

Grok 4.20 Multi-Agent Beta (0309)

1,447.71 rating

Confidence interval: 1,433.47 – 1,461.95

1995 votes

Oct 28, 2026

Category: spanish
Evaluated configuration: Grok 4.20 Multi-Agent Beta (0309)
Evaluation type: human preference

Run it locally

Inference · memory estimate

Grok 4.20 Multi-Agent Beta (0309)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license: Proprietary