GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

reka-core-20240904

Model evaluation

Capability & preference

Human preference · Arena

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

reka-core-20240904 1248.3 reka-core-20240904 · 95% interval 1241.2–1255.4 · 7,312 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

reka-core-20240904

1,241.02 rating

Confidence interval: 1,218.02 – 1,264.02

598 votes

Category
chinese
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

reka-core-20240904

1,229.84 rating

Confidence interval: 1,214.46 – 1,245.22

1216 votes

Category
coding
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

reka-core-20240904

1,236.09 rating

Confidence interval: 1,218.10 – 1,254.08

1037 votes

Category
creative_writing
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

reka-core-20240904

1,255.97 rating

Confidence interval: 1,246.67 – 1,265.26

3934 votes

Category
english
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

reka-core-20240904

1,156.33 rating

Confidence interval: 1,144.74 – 1,167.92

4480 votes

Category
exclude_ties
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

reka-core-20240904

1,213.36 rating

Confidence interval: 1,188.65 – 1,238.07

458 votes

Category
expert
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

reka-core-20240904

1,243.19 rating

Confidence interval: 1,202.79 – 1,283.60

183 votes

Category
german
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

reka-core-20240904

1,210.83 rating

Confidence interval: 1,199.32 – 1,222.35

2145 votes

Category
hard_prompts
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

reka-core-20240904

1,215.45 rating

Confidence interval: 1,201.22 – 1,229.68

1341 votes

Category
hard_prompts_english
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

reka-core-20240904

1,217.10 rating

Confidence interval: 1,197.59 – 1,236.60

837 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

reka-core-20240904

1,207.22 rating

Confidence interval: 1,190.27 – 1,224.17

1092 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

reka-core-20240904

1,273.24 rating

Confidence interval: 1,244.59 – 1,301.89

371 votes

Category
industry_legal_and_government
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

reka-core-20240904

1,265.43 rating

Confidence interval: 1,249.26 – 1,281.59

1183 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

reka-core-20240904

1,239.01 rating

Confidence interval: 1,223.35 – 1,254.67

1039 votes

Category
industry_mathematical
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

reka-core-20240904

1,278.14 rating

Confidence interval: 1,248.65 – 1,307.63

364 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

reka-core-20240904

1,231.68 rating

Confidence interval: 1,219.12 – 1,244.23

2006 votes

Category
industry_software_and_it_services
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

reka-core-20240904

1,242.40 rating

Confidence interval: 1,229.06 – 1,255.74

1890 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

reka-core-20240904

1,211.74 rating

Confidence interval: 1,201.69 – 1,221.80

3118 votes

Category
instruction_following
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

reka-core-20240904

1,220.12 rating

Confidence interval: 1,202.01 – 1,238.23

951 votes

Category
longer_query
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

reka-core-20240904

1,221.63 rating

Confidence interval: 1,207.30 – 1,235.97

1207 votes

Category
math
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

reka-core-20240904

1,226.56 rating

Confidence interval: 1,210.38 – 1,242.75

1255 votes

Category
multi_turn
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

reka-core-20240904

1,236.50 rating

Confidence interval: 1,226.55 – 1,246.44

3378 votes

Category
non_english
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

reka-core-20240904

1,248.34 rating

Confidence interval: 1,241.24 – 1,255.45

7312 votes

Category
overall
Evaluated configuration
reka-core-20240904
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

reka-core-20240904

1,249.21 rating

Confidence interval: 1,231.44 – 1,266.98

976 votes

Category
russian
Evaluated configuration
reka-core-20240904
Evaluation type
human preference

Run it locally

Inference · memory estimate

reka-core-20240904

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary