GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

DBRX Instruct Preview

Model evaluation

Capability & preference

Human preference · Arena

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

DBRX Instruct Preview 1118.8 DBRX Instruct Preview · 95% interval 1112.8–1124.9 · 32,191 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

DBRX Instruct Preview

1,067.76 rating

Confidence interval: 1,054.69 – 1,080.83

5019 votes

Category
chinese
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,131.47 rating

Confidence interval: 1,120.61 – 1,142.32

5502 votes

Category
coding
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,103.94 rating

Confidence interval: 1,092.39 – 1,115.49

4808 votes

Category
creative_writing
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,155.07 rating

Confidence interval: 1,147.50 – 1,162.63

18053 votes

Category
english
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

DBRX Instruct Preview

958.68 rating

Confidence interval: 949.42 – 967.95

21404 votes

Category
exclude_ties
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,075.89 rating

Confidence interval: 1,059.73 – 1,092.05

1678 votes

Category
expert
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,096.37 rating

Confidence interval: 1,065.94 – 1,126.81

538 votes

Category
french
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,057.36 rating

Confidence interval: 1,033.93 – 1,080.79

742 votes

Category
german
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,112.85 rating

Confidence interval: 1,103.87 – 1,121.84

8865 votes

Category
hard_prompts
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,137.61 rating

Confidence interval: 1,126.91 – 1,148.32

5516 votes

Category
hard_prompts_english
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,072.05 rating

Confidence interval: 1,058.79 – 1,085.31

3376 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,098.72 rating

Confidence interval: 1,087.37 – 1,110.06

4630 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,104.54 rating

Confidence interval: 1,087.75 – 1,121.33

1748 votes

Category
industry_legal_and_government
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,099.90 rating

Confidence interval: 1,088.25 – 1,111.54

5642 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,130.99 rating

Confidence interval: 1,118.88 – 1,143.10

3599 votes

Category
industry_mathematical
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,067.80 rating

Confidence interval: 1,050.11 – 1,085.49

1705 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,116.40 rating

Confidence interval: 1,107.13 – 1,125.67

9343 votes

Category
industry_software_and_it_services
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,121.43 rating

Confidence interval: 1,111.69 – 1,131.17

7999 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,111.43 rating

Confidence interval: 1,103.01 – 1,119.85

11274 votes

Category
instruction_following
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

DBRX Instruct Preview

989.68 rating

Confidence interval: 958.78 – 1,020.58

426 votes

Category
japanese
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

992.80 rating

Confidence interval: 961.71 – 1,023.89

442 votes

Category
korean
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,111.30 rating

Confidence interval: 1,098.91 – 1,123.69

3489 votes

Category
longer_query
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,145.02 rating

Confidence interval: 1,133.54 – 1,156.49

4001 votes

Category
math
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,111.00 rating

Confidence interval: 1,099.31 – 1,122.68

4918 votes

Category
multi_turn
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,071.13 rating

Confidence interval: 1,063.21 – 1,079.04

14138 votes

Category
non_english
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,118.84 rating

Confidence interval: 1,112.80 – 1,124.89

32191 votes

Category
overall
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,077.78 rating

Confidence interval: 1,064.81 – 1,090.75

3044 votes

Category
russian
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

DBRX Instruct Preview

1,064.00 rating

Confidence interval: 1,031.19 – 1,096.81

404 votes

Category
spanish
Evaluated configuration
DBRX Instruct Preview
Evaluation type
human preference

Run it locally

Inference · memory estimate

DBRX Instruct Preview

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
DBRX LICENSE