GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

Vicuna 33B

Model evaluation

Capability & preference

Human preference · Arena

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

Vicuna 33B 1105.2 Vicuna 33B · 95% interval 1099.1–1111.3 · 22,479 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 33B

1,035.80 rating

Confidence interval: 1,018.61 – 1,052.99

1385 votes

Category
chinese
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Vicuna 33B

1,077.96 rating

Confidence interval: 1,065.36 – 1,090.55

2866 votes

Category
coding
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Vicuna 33B

1,124.38 rating

Confidence interval: 1,112.51 – 1,136.24

4010 votes

Category
creative_writing
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Vicuna 33B

1,133.49 rating

Confidence interval: 1,126.26 – 1,140.72

16692 votes

Category
english
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Vicuna 33B

944.63 rating

Confidence interval: 935.55 – 953.72

15324 votes

Category
exclude_ties
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Vicuna 33B

1,029.00 rating

Confidence interval: 1,002.29 – 1,055.71

484 votes

Category
expert
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Vicuna 33B

1,086.07 rating

Confidence interval: 1,048.02 – 1,124.11

295 votes

Category
french
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Vicuna 33B

1,059.12 rating

Confidence interval: 1,029.50 – 1,088.75

428 votes

Category
german
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Vicuna 33B

1,066.74 rating

Confidence interval: 1,056.86 – 1,076.63

5071 votes

Category
hard_prompts
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Vicuna 33B

1,079.56 rating

Confidence interval: 1,068.38 – 1,090.74

3964 votes

Category
hard_prompts_english
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Vicuna 33B

1,054.34 rating

Confidence interval: 1,038.91 – 1,069.77

1909 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Vicuna 33B

1,102.83 rating

Confidence interval: 1,090.91 – 1,114.75

3642 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Vicuna 33B

1,098.85 rating

Confidence interval: 1,079.55 – 1,118.16

1246 votes

Category
industry_legal_and_government
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Vicuna 33B

1,099.69 rating

Confidence interval: 1,088.17 – 1,111.22

3986 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Vicuna 33B

1,076.67 rating

Confidence interval: 1,063.16 – 1,090.17

2352 votes

Category
industry_mathematical
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Vicuna 33B

1,060.15 rating

Confidence interval: 1,041.00 – 1,079.29

1203 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Vicuna 33B

1,085.74 rating

Confidence interval: 1,075.69 – 1,095.79

5525 votes

Category
industry_software_and_it_services
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Vicuna 33B

1,118.74 rating

Confidence interval: 1,108.68 – 1,128.79

5627 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Vicuna 33B

1,064.82 rating

Confidence interval: 1,055.93 – 1,073.70

6983 votes

Category
instruction_following
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Vicuna 33B

958.70 rating

Confidence interval: 916.89 – 1,000.52

195 votes

Category
japanese
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Vicuna 33B

1,028.34 rating

Confidence interval: 1,009.41 – 1,047.27

1047 votes

Category
longer_query
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Vicuna 33B

1,070.63 rating

Confidence interval: 1,058.09 – 1,083.17

2663 votes

Category
math
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Vicuna 33B

1,080.45 rating

Confidence interval: 1,067.57 – 1,093.34

3114 votes

Category
multi_turn
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Vicuna 33B

1,054.76 rating

Confidence interval: 1,045.45 – 1,064.08

5787 votes

Category
non_english
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Vicuna 33B

1,105.23 rating

Confidence interval: 1,099.13 – 1,111.34

22479 votes

Category
overall
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Vicuna 33B

1,082.48 rating

Confidence interval: 1,060.59 – 1,104.38

703 votes

Category
russian
Evaluated configuration
Vicuna 33B
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Vicuna 33B

1,079.89 rating

Confidence interval: 1,041.91 – 1,117.87

275 votes

Category
spanish
Evaluated configuration
Vicuna 33B
Evaluation type
human preference

Run it locally

Inference · memory estimate

Vicuna 33B

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Non-commercial