GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

athene-70b-0725

Model evaluation

Capability & preference

Human preference · Arena

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

athene-70b-0725 1265.1 athene-70b-0725 · 95% interval 1259.5–1270.7 · 19,621 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

athene-70b-0725

1,237.53 rating

Confidence interval: 1,222.89 – 1,252.18

1678 votes

Category
chinese
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

athene-70b-0725

1,266.77 rating

Confidence interval: 1,255.99 – 1,277.55

3122 votes

Category
coding
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

athene-70b-0725

1,238.16 rating

Confidence interval: 1,226.92 – 1,249.41

3097 votes

Category
creative_writing
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

athene-70b-0725

1,287.97 rating

Confidence interval: 1,281.04 – 1,294.90

11268 votes

Category
english
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

athene-70b-0725

1,187.70 rating

Confidence interval: 1,178.97 – 1,196.44

12348 votes

Category
exclude_ties
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

athene-70b-0725

1,227.56 rating

Confidence interval: 1,208.32 – 1,246.81

867 votes

Category
expert
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

athene-70b-0725

1,286.46 rating

Confidence interval: 1,243.06 – 1,329.86

199 votes

Category
french
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

athene-70b-0725

1,251.64 rating

Confidence interval: 1,227.71 – 1,275.57

585 votes

Category
german
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

athene-70b-0725

1,242.97 rating

Confidence interval: 1,234.41 – 1,251.54

5448 votes

Category
hard_prompts
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

athene-70b-0725

1,254.53 rating

Confidence interval: 1,244.48 – 1,264.59

3605 votes

Category
hard_prompts_english
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

athene-70b-0725

1,245.54 rating

Confidence interval: 1,232.19 – 1,258.90

2092 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

athene-70b-0725

1,232.54 rating

Confidence interval: 1,222.02 – 1,243.07

3246 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

athene-70b-0725

1,292.95 rating

Confidence interval: 1,273.81 – 1,312.08

1008 votes

Category
industry_legal_and_government
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

athene-70b-0725

1,293.67 rating

Confidence interval: 1,282.19 – 1,305.16

3029 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

athene-70b-0725

1,233.23 rating

Confidence interval: 1,222.14 – 1,244.32

2559 votes

Category
industry_mathematical
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

athene-70b-0725

1,272.21 rating

Confidence interval: 1,253.26 – 1,291.15

1039 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

athene-70b-0725

1,265.18 rating

Confidence interval: 1,256.02 – 1,274.34

5037 votes

Category
industry_software_and_it_services
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

athene-70b-0725

1,240.83 rating

Confidence interval: 1,231.97 – 1,249.69

5378 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

athene-70b-0725

1,226.66 rating

Confidence interval: 1,218.97 – 1,234.35

7490 votes

Category
instruction_following
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

athene-70b-0725

1,138.22 rating

Confidence interval: 1,114.94 – 1,161.50

676 votes

Category
japanese
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

athene-70b-0725

1,150.66 rating

Confidence interval: 1,118.71 – 1,182.61

396 votes

Category
korean
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

athene-70b-0725

1,231.89 rating

Confidence interval: 1,219.56 – 1,244.21

2123 votes

Category
longer_query
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

athene-70b-0725

1,231.30 rating

Confidence interval: 1,220.85 – 1,241.75

2921 votes

Category
math
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

athene-70b-0725

1,257.20 rating

Confidence interval: 1,246.35 – 1,268.04

3344 votes

Category
multi_turn
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

athene-70b-0725

1,234.71 rating

Confidence interval: 1,227.13 – 1,242.29

8353 votes

Category
non_english
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

athene-70b-0725

1,265.10 rating

Confidence interval: 1,259.46 – 1,270.74

19621 votes

Category
overall
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

athene-70b-0725

1,250.39 rating

Confidence interval: 1,237.21 – 1,263.58

1934 votes

Category
russian
Evaluated configuration
athene-70b-0725
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

athene-70b-0725

1,259.17 rating

Confidence interval: 1,223.56 – 1,294.79

271 votes

Category
spanish
Evaluated configuration
athene-70b-0725
Evaluation type
human preference

Run it locally

Inference · memory estimate

athene-70b-0725

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
CC-BY-NC-4.0