GeekRadar / The comparison desk
COMPARE.
AI models
Not reported

Dolphin 2.2.1 Mistral 7B

Model evaluation

Capability & preference

Human preference · Arena

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

Dolphin 2.2.1 Mistral 7B 1080.9 Dolphin 2.2.1 Mistral 7B · 95% interval 1065.7–1096.2 · 1,679 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 / creative_writinghuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,085.44 rating

Confidence interval: 1,052.32 – 1,118.55

291 votes

Category
creative_writing
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,112.46 rating

Confidence interval: 1,095.77 – 1,129.15

1302 votes

Category
english
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

905.99 rating

Confidence interval: 884.02 – 927.97

1098 votes

Category
exclude_ties
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,054.44 rating

Confidence interval: 1,027.42 – 1,081.46

359 votes

Category
hard_prompts
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,068.05 rating

Confidence interval: 1,037.20 – 1,098.89

286 votes

Category
hard_prompts_english
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,049.51 rating

Confidence interval: 1,016.28 – 1,082.74

282 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,065.35 rating

Confidence interval: 1,032.62 – 1,098.08

308 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,055.29 rating

Confidence interval: 1,021.20 – 1,089.37

201 votes

Category
industry_mathematical
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,046.42 rating

Confidence interval: 1,018.56 – 1,074.28

366 votes

Category
industry_software_and_it_services
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,085.88 rating

Confidence interval: 1,057.51 – 1,114.25

382 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,053.21 rating

Confidence interval: 1,028.96 – 1,077.46

497 votes

Category
instruction_following
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,076.59 rating

Confidence interval: 1,044.23 – 1,108.94

219 votes

Category
math
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,020.60 rating

Confidence interval: 980.31 – 1,060.90

208 votes

Category
multi_turn
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,012.34 rating

Confidence interval: 985.02 – 1,039.66

377 votes

Category
non_english
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Dolphin 2.2.1 Mistral 7B

1,080.95 rating

Confidence interval: 1,065.67 – 1,096.23

1679 votes

Category
overall
Evaluated configuration
Dolphin 2.2.1 Mistral 7B
Evaluation type
human preference

Run it locally

Inference · memory estimate

Dolphin 2.2.1 Mistral 7B

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Apache-2.0