GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

OpenAssistant Pythia 12B

Model evaluation

Capability & preference

Human preference · Arena

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

OpenAssistant Pythia 12B 916.0 OpenAssistant Pythia 12B · 95% interval 905.4–926.7 · 6,310 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

OpenAssistant Pythia 12B

802.90 rating

Confidence interval: 770.32 – 835.49

321 votes

Category
chinese
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

900.03 rating

Confidence interval: 875.56 – 924.51

714 votes

Category
coding
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

922.99 rating

Confidence interval: 903.09 – 942.90

1094 votes

Category
creative_writing
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

945.40 rating

Confidence interval: 933.73 – 957.08

5212 votes

Category
english
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

671.52 rating

Confidence interval: 655.64 – 687.40

4325 votes

Category
exclude_ties
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

884.45 rating

Confidence interval: 865.92 – 902.98

1153 votes

Category
hard_prompts
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

897.49 rating

Confidence interval: 877.28 – 917.70

1002 votes

Category
hard_prompts_english
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

822.35 rating

Confidence interval: 792.27 – 852.43

454 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

885.22 rating

Confidence interval: 864.56 – 905.88

1067 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

922.45 rating

Confidence interval: 882.28 – 962.62

266 votes

Category
industry_legal_and_government
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

896.60 rating

Confidence interval: 875.93 – 917.26

1134 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

890.65 rating

Confidence interval: 865.78 – 915.52

629 votes

Category
industry_mathematical
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

819.01 rating

Confidence interval: 771.91 – 866.11

201 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

881.26 rating

Confidence interval: 862.70 – 899.81

1409 votes

Category
industry_software_and_it_services
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

928.15 rating

Confidence interval: 910.65 – 945.65

1408 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

887.23 rating

Confidence interval: 871.69 – 902.76

1744 votes

Category
instruction_following
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

891.86 rating

Confidence interval: 870.04 – 913.67

687 votes

Category
math
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

869.24 rating

Confidence interval: 844.19 – 894.29

683 votes

Category
multi_turn
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

848.01 rating

Confidence interval: 830.22 – 865.81

1098 votes

Category
non_english
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

OpenAssistant Pythia 12B

916.04 rating

Confidence interval: 905.36 – 926.71

6310 votes

Category
overall
Evaluated configuration
OpenAssistant Pythia 12B
Evaluation type
human preference

Run it locally

Inference · memory estimate

OpenAssistant Pythia 12B

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Apache 2.0