GeekRadar / The comparison desk
COMPARE.
AI models
meta

Llama-4-Maverick-17B-128E-Instruct

AI model

Open weights
Yes

Capability & preference

Human preference · Arena

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

Llama-4-Maverick-17B-128E-Instruct 1287.2 Llama 4 Maverick 17B 128E Instruct · 95% interval 1283.0–1291.5 · 39,439 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

Llama-4-Maverick-17B-128E-Instruct

1,276.74 rating

Confidence interval: 1,264.56 – 1,288.91

2597 votes

Category
chinese
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,301.26 rating

Confidence interval: 1,293.80 – 1,308.71

6937 votes

Category
coding
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,265.97 rating

Confidence interval: 1,257.43 – 1,274.52

5150 votes

Category
creative_writing
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,302.56 rating

Confidence interval: 1,297.22 – 1,307.90

18916 votes

Category
english
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,220.47 rating

Confidence interval: 1,214.52 – 1,226.42

28104 votes

Category
exclude_ties
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,258.85 rating

Confidence interval: 1,245.52 – 1,272.19

2016 votes

Category
expert
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,259.83 rating

Confidence interval: 1,234.38 – 1,285.28

596 votes

Category
french
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,290.88 rating

Confidence interval: 1,271.87 – 1,309.90

1029 votes

Category
german
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,280.58 rating

Confidence interval: 1,274.88 – 1,286.29

15866 votes

Category
hard_prompts
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,295.49 rating

Confidence interval: 1,288.21 – 1,302.76

7691 votes

Category
hard_prompts_english
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,260.59 rating

Confidence interval: 1,252.48 – 1,268.71

6124 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,249.77 rating

Confidence interval: 1,242.21 – 1,257.32

6925 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,285.75 rating

Confidence interval: 1,273.62 – 1,297.88

2583 votes

Category
industry_legal_and_government
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,287.56 rating

Confidence interval: 1,279.70 – 1,295.43

6714 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,294.45 rating

Confidence interval: 1,282.92 – 1,305.99

2568 votes

Category
industry_mathematical
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,275.39 rating

Confidence interval: 1,262.19 – 1,288.59

2294 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,298.65 rating

Confidence interval: 1,292.52 – 1,304.79

12285 votes

Category
industry_software_and_it_services
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,277.22 rating

Confidence interval: 1,270.33 – 1,284.11

8729 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,266.89 rating

Confidence interval: 1,260.43 – 1,273.34

10498 votes

Category
instruction_following
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,206.38 rating

Confidence interval: 1,185.15 – 1,227.61

935 votes

Category
japanese
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,203.29 rating

Confidence interval: 1,181.36 – 1,225.22

796 votes

Category
korean
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,279.56 rating

Confidence interval: 1,272.11 – 1,287.01

7074 votes

Category
longer_query
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,299.32 rating

Confidence interval: 1,288.51 – 1,310.13

2825 votes

Category
math
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,289.36 rating

Confidence interval: 1,281.53 – 1,297.18

6680 votes

Category
multi_turn
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,268.97 rating

Confidence interval: 1,263.72 – 1,274.23

20513 votes

Category
non_english
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,287.21 rating

Confidence interval: 1,282.96 – 1,291.47

39439 votes

Category
overall
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,275.52 rating

Confidence interval: 1,262.99 – 1,288.05

2675 votes

Category
polish
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,285.85 rating

Confidence interval: 1,275.42 – 1,296.27

3102 votes

Category
russian
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Llama-4-Maverick-17B-128E-Instruct

1,293.19 rating

Confidence interval: 1,271.77 – 1,314.60

820 votes

Category
spanish
Evaluated configuration
Llama 4 Maverick 17B 128E Instruct
Evaluation type
human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,276.93 rating

Confidence interval: 1,264.77 – 1,289.09

2599 votes

Category
chinese
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,301.40 rating

Confidence interval: 1,293.95 – 1,308.86

6940 votes

Category
coding
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,266.22 rating

Confidence interval: 1,257.68 – 1,274.77

5151 votes

Category
creative_writing
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,302.50 rating

Confidence interval: 1,297.16 – 1,307.84

18917 votes

Category
english
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,220.48 rating

Confidence interval: 1,214.53 – 1,226.43

28112 votes

Category
exclude_ties
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,258.89 rating

Confidence interval: 1,245.56 – 1,272.23

2016 votes

Category
expert
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,259.41 rating

Confidence interval: 1,233.95 – 1,284.86

596 votes

Category
french
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,291.16 rating

Confidence interval: 1,272.16 – 1,310.16

1028 votes

Category
german
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,280.59 rating

Confidence interval: 1,274.88 – 1,286.29

15870 votes

Category
hard_prompts
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,295.35 rating

Confidence interval: 1,288.08 – 1,302.63

7692 votes

Category
hard_prompts_english
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,260.59 rating

Confidence interval: 1,252.48 – 1,268.71

6128 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,249.90 rating

Confidence interval: 1,242.34 – 1,257.45

6927 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,285.63 rating

Confidence interval: 1,273.50 – 1,297.75

2583 votes

Category
industry_legal_and_government
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,287.60 rating

Confidence interval: 1,279.73 – 1,295.46

6713 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,294.53 rating

Confidence interval: 1,283.01 – 1,306.06

2569 votes

Category
industry_mathematical
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,275.40 rating

Confidence interval: 1,262.20 – 1,288.60

2293 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,298.70 rating

Confidence interval: 1,292.57 – 1,304.83

12287 votes

Category
industry_software_and_it_services
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,277.37 rating

Confidence interval: 1,270.48 – 1,284.26

8730 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,266.95 rating

Confidence interval: 1,260.50 – 1,273.41

10500 votes

Category
instruction_following
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,205.94 rating

Confidence interval: 1,184.74 – 1,227.15

936 votes

Category
japanese
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,203.68 rating

Confidence interval: 1,181.78 – 1,225.58

799 votes

Category
korean
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,279.67 rating

Confidence interval: 1,272.22 – 1,287.11

7075 votes

Category
longer_query
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,299.33 rating

Confidence interval: 1,288.52 – 1,310.14

2825 votes

Category
math
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,289.38 rating

Confidence interval: 1,281.56 – 1,297.21

6681 votes

Category
multi_turn
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,269.08 rating

Confidence interval: 1,263.83 – 1,274.33

20523 votes

Category
non_english
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,287.19 rating

Confidence interval: 1,282.93 – 1,291.44

39450 votes

Category
overall
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,275.58 rating

Confidence interval: 1,263.06 – 1,288.11

2675 votes

Category
polish
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,286.12 rating

Confidence interval: 1,275.70 – 1,296.55

3103 votes

Category
russian
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

Llama-4-Maverick-17B-128E-Instruct

1,293.10 rating

Confidence interval: 1,271.68 – 1,314.52

820 votes

Category
spanish
Evaluated configuration
llama-4-maverick-17b-128e-instruct
Evaluation type
human preference

Run it locally

Inference · memory estimate

Llama-4-Maverick-17B-128E-Instruct

748.0 GiB checkpoint tensor payload · 401.58 B stored tensor elements · BF16

Stored tensors can include buffers and repeated shared weights. A verified unique parameter count is unavailable. Checkpoint size is not the memory required to run the model.

Capabilities

Open weights
Yes
Model license
other
Checkpoint tensor payload
748.01 GiB

Input modalities

  • text
  • image

Output modalities

  • text