GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

Vicuna 7B

Model evaluation

Capability & preference

Human preference · Arena

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

Vicuna 7B 1030.9 Vicuna 7B · 95% interval 1021.9–1040.0 · 6,923 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

Vicuna 7B

978.97 rating

Confidence interval: 946.17 – 1,011.77

335 votes

Category
chinese
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Vicuna 7B

1,009.74 rating

Confidence interval: 987.36 – 1,032.11

726 votes

Category
coding
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Vicuna 7B

1,020.78 rating

Confidence interval: 1,002.96 – 1,038.60

1323 votes

Category
creative_writing
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Vicuna 7B

1,051.30 rating

Confidence interval: 1,041.43 – 1,061.17

5577 votes

Category
english
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Vicuna 7B

840.87 rating

Confidence interval: 827.72 – 854.02

4750 votes

Category
exclude_ties
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Vicuna 7B

994.76 rating

Confidence interval: 978.14 – 1,011.38

1258 votes

Category
hard_prompts
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Vicuna 7B

1,004.35 rating

Confidence interval: 986.00 – 1,022.69

1066 votes

Category
hard_prompts_english
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Vicuna 7B

972.60 rating

Confidence interval: 945.30 – 999.89

555 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Vicuna 7B

1,016.49 rating

Confidence interval: 999.12 – 1,033.87

1285 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Vicuna 7B

1,032.06 rating

Confidence interval: 997.13 – 1,067.00

305 votes

Category
industry_legal_and_government
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Vicuna 7B

1,000.89 rating

Confidence interval: 982.15 – 1,019.63

1171 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Vicuna 7B

979.96 rating

Confidence interval: 956.31 – 1,003.60

590 votes

Category
industry_mathematical
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Vicuna 7B

997.72 rating

Confidence interval: 962.03 – 1,033.41

291 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Vicuna 7B

1,017.03 rating

Confidence interval: 1,000.93 – 1,033.12

1601 votes

Category
industry_software_and_it_services
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Vicuna 7B

1,048.51 rating

Confidence interval: 1,033.53 – 1,063.48

1783 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Vicuna 7B

993.65 rating

Confidence interval: 979.78 – 1,007.52

2020 votes

Category
instruction_following
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Vicuna 7B

1,035.24 rating

Confidence interval: 992.49 – 1,078.00

162 votes

Category
longer_query
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Vicuna 7B

993.79 rating

Confidence interval: 972.42 – 1,015.17

658 votes

Category
math
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Vicuna 7B

987.81 rating

Confidence interval: 966.90 – 1,008.72

935 votes

Category
multi_turn
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Vicuna 7B

1,002.83 rating

Confidence interval: 986.22 – 1,019.44

1346 votes

Category
non_english
Evaluated configuration
Vicuna 7B
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Vicuna 7B

1,030.92 rating

Confidence interval: 1,021.85 – 1,039.99

6923 votes

Category
overall
Evaluated configuration
Vicuna 7B
Evaluation type
human preference

Run it locally

Inference · memory estimate

Vicuna 7B

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Llama 2 Community