GeekRadar / The comparison desk

COMPARE.

AI models

Code Llama 34B Instruct

Model evaluation

Capability & preference

Human preference · Arena

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

Code Llama 34B Instruct 1065.5 Code Llama 34B Instruct · 95% interval 1056.7–1074.2 · 7,366 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

Code Llama 34B Instruct

976.39 rating

Confidence interval: 938.83 – 1,013.96

249 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / codinghuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,045.52 rating

Confidence interval: 1,025.41 – 1,065.62

853 votes

Oct 28, 2026

Category: coding
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,031.30 rating

Confidence interval: 1,014.69 – 1,047.91

1507 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / englishhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,091.99 rating

Confidence interval: 1,082.35 – 1,101.63

5804 votes

Oct 28, 2026

Category: english
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

882.57 rating

Confidence interval: 870.00 – 895.14

5094 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,031.96 rating

Confidence interval: 1,017.44 – 1,046.48

1540 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,046.91 rating

Confidence interval: 1,030.99 – 1,062.82

1278 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,033.56 rating

Confidence interval: 1,009.17 – 1,057.96

639 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,014.61 rating

Confidence interval: 997.83 – 1,031.38

1375 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,064.30 rating

Confidence interval: 1,031.92 – 1,096.68

383 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,066.44 rating

Confidence interval: 1,048.85 – 1,084.02

1225 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,054.97 rating

Confidence interval: 1,033.79 – 1,076.15

667 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,025.76 rating

Confidence interval: 993.86 – 1,057.67

340 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,062.81 rating

Confidence interval: 1,048.01 – 1,077.61

1769 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,048.23 rating

Confidence interval: 1,033.74 – 1,062.73

2003 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,027.76 rating

Confidence interval: 1,014.98 – 1,040.55

2294 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,013.09 rating

Confidence interval: 980.63 – 1,045.56

320 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / mathhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,056.41 rating

Confidence interval: 1,037.28 – 1,075.55

770 votes

Oct 28, 2026

Category: math
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,015.35 rating

Confidence interval: 996.99 – 1,033.71

1155 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,010.48 rating

Confidence interval: 994.85 – 1,026.12

1562 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Arena text preference / overallhuman preference · 2026-10-28 · rating

Code Llama 34B Instruct

1,065.46 rating

Confidence interval: 1,056.70 – 1,074.22

7366 votes

Oct 28, 2026

Category: overall
Evaluated configuration: Code Llama 34B Instruct
Evaluation type: human preference

Run it locally

Inference · memory estimate

Code Llama 34B Instruct

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license: Llama 2 Community