GeekRadar / The comparison desk
COMPARE.
AI models
meta

llama-3.1-405b-instruct-bf16

Model evaluation

Capability & preference

Human preference · Arena

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

llama-3.1-405b-instruct-bf16 1283.7 llama-3.1-405b-instruct-bf16 · 95% interval 1280.0–1287.3 · 41,375 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

llama-3.1-405b-instruct-bf16

1,241.86 rating

Confidence interval: 1,230.82 – 1,252.91

3105 votes

Category
chinese
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,291.14 rating

Confidence interval: 1,283.67 – 1,298.61

6249 votes

Category
coding
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,259.58 rating

Confidence interval: 1,251.94 – 1,267.21

6512 votes

Category
creative_writing
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,311.77 rating

Confidence interval: 1,307.17 – 1,316.36

23318 votes

Category
english
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,213.36 rating

Confidence interval: 1,207.79 – 1,218.93

27059 votes

Category
exclude_ties
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,228.80 rating

Confidence interval: 1,215.95 – 1,241.65

2128 votes

Category
expert
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,278.68 rating

Confidence interval: 1,248.03 – 1,309.33

364 votes

Category
french
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,252.34 rating

Confidence interval: 1,232.92 – 1,271.76

894 votes

Category
german
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,269.27 rating

Confidence interval: 1,263.33 – 1,275.21

10652 votes

Category
hard_prompts
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,291.24 rating

Confidence interval: 1,283.97 – 1,298.50

6791 votes

Category
hard_prompts_english
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,244.88 rating

Confidence interval: 1,235.76 – 1,254.00

4512 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,240.93 rating

Confidence interval: 1,233.66 – 1,248.20

7230 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,301.61 rating

Confidence interval: 1,288.78 – 1,314.43

2330 votes

Category
industry_legal_and_government
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,295.07 rating

Confidence interval: 1,287.72 – 1,302.42

7177 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,275.63 rating

Confidence interval: 1,266.93 – 1,284.32

4366 votes

Category
industry_mathematical
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,286.38 rating

Confidence interval: 1,272.37 – 1,300.39

1899 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,290.49 rating

Confidence interval: 1,284.29 – 1,296.69

10298 votes

Category
industry_software_and_it_services
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,270.52 rating

Confidence interval: 1,264.38 – 1,276.65

11150 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,258.46 rating

Confidence interval: 1,253.22 – 1,263.71

16174 votes

Category
instruction_following
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,207.89 rating

Confidence interval: 1,187.28 – 1,228.50

846 votes

Category
japanese
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,172.36 rating

Confidence interval: 1,146.41 – 1,198.32

586 votes

Category
korean
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,265.47 rating

Confidence interval: 1,257.48 – 1,273.46

5486 votes

Category
longer_query
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,278.19 rating

Confidence interval: 1,270.16 – 1,286.22

5215 votes

Category
math
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,296.85 rating

Confidence interval: 1,289.20 – 1,304.50

6797 votes

Category
multi_turn
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,246.82 rating

Confidence interval: 1,241.77 – 1,251.88

18057 votes

Category
non_english
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,283.65 rating

Confidence interval: 1,279.96 – 1,287.35

41375 votes

Category
overall
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,256.11 rating

Confidence interval: 1,247.43 – 1,264.79

4863 votes

Category
russian
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

llama-3.1-405b-instruct-bf16

1,259.75 rating

Confidence interval: 1,229.39 – 1,290.12

348 votes

Category
spanish
Evaluated configuration
llama-3.1-405b-instruct-bf16
Evaluation type
human preference

Run it locally

Inference · memory estimate

llama-3.1-405b-instruct-bf16

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Llama 3.1 Community