GeekRadar / The comparison desk
COMPARE.
AI models
mistral

mixtral-8x22b-instruct-v0.1

Model evaluation

Capability & preference

Human preference · Arena

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

mixtral-8x22b-instruct-v0.1 1161.9 mixtral-8x22b-instruct-v0.1 · 95% interval 1157.4–1166.5 · 51,416 votes · 2026-10-15

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 · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,116.35 rating

Confidence interval: 1,105.22 – 1,127.47

5655 votes

Category
chinese
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,165.42 rating

Confidence interval: 1,156.87 – 1,173.96

8780 votes

Category
coding
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,141.21 rating

Confidence interval: 1,131.96 – 1,150.47

7447 votes

Category
creative_writing
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,187.09 rating

Confidence interval: 1,181.45 – 1,192.72

29892 votes

Category
english
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,023.70 rating

Confidence interval: 1,016.74 – 1,030.67

33717 votes

Category
exclude_ties
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,113.00 rating

Confidence interval: 1,099.33 – 1,126.66

2582 votes

Category
expert
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,166.05 rating

Confidence interval: 1,140.09 – 1,192.01

767 votes

Category
french
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,141.23 rating

Confidence interval: 1,123.90 – 1,158.55

1653 votes

Category
german
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,149.59 rating

Confidence interval: 1,142.67 – 1,156.51

14707 votes

Category
hard_prompts
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,162.83 rating

Confidence interval: 1,154.58 – 1,171.09

9466 votes

Category
hard_prompts_english
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,118.60 rating

Confidence interval: 1,108.48 – 1,128.72

5554 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,138.64 rating

Confidence interval: 1,129.74 – 1,147.54

7616 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,164.86 rating

Confidence interval: 1,151.31 – 1,178.41

2798 votes

Category
industry_legal_and_government
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,150.94 rating

Confidence interval: 1,141.92 – 1,159.95

8677 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,179.00 rating

Confidence interval: 1,169.49 – 1,188.50

6001 votes

Category
industry_mathematical
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,119.79 rating

Confidence interval: 1,106.08 – 1,133.50

2823 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,154.11 rating

Confidence interval: 1,146.82 – 1,161.40

14389 votes

Category
industry_software_and_it_services
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,161.04 rating

Confidence interval: 1,153.48 – 1,168.60

12808 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,146.99 rating

Confidence interval: 1,140.52 – 1,153.46

18515 votes

Category
instruction_following
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,036.62 rating

Confidence interval: 1,015.06 – 1,058.17

972 votes

Category
japanese
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,056.64 rating

Confidence interval: 1,034.61 – 1,078.67

1187 votes

Category
korean
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,143.50 rating

Confidence interval: 1,133.75 – 1,153.24

5865 votes

Category
longer_query
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,184.23 rating

Confidence interval: 1,175.28 – 1,193.18

6778 votes

Category
math
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,130.15 rating

Confidence interval: 1,121.16 – 1,139.13

7764 votes

Category
multi_turn
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,128.17 rating

Confidence interval: 1,121.95 – 1,134.39

21524 votes

Category
non_english
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,161.95 rating

Confidence interval: 1,157.43 – 1,166.46

51416 votes

Category
overall
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,171.01 rating

Confidence interval: 1,110.45 – 1,231.58

132 votes

Category
polish
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,157.72 rating

Confidence interval: 1,147.51 – 1,167.93

4914 votes

Category
russian
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

mixtral-8x22b-instruct-v0.1

1,150.57 rating

Confidence interval: 1,125.03 – 1,176.12

826 votes

Category
spanish
Evaluated configuration
mixtral-8x22b-instruct-v0.1
Evaluation type
human preference

Run it locally

Inference · memory estimate

mixtral-8x22b-instruct-v0.1

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Apache 2.0