GeekRadar / The comparison desk

COMPARE.

AI models

microsoft

WizardLM 70B

Model evaluation

Capability & preference

Human preference · Arena

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

WizardLM 70B 1119.5 WizardLM 70B · 95% interval 1110.2–1128.8 · 8,214 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

WizardLM 70B

1,052.35 rating

Confidence interval: 1,020.14 – 1,084.57

334 votes

Oct 28, 2026

Category: chinese
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,080.87 rating

Confidence interval: 1,061.27 – 1,100.47

988 votes

Oct 28, 2026

Category: coding
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,148.89 rating

Confidence interval: 1,132.17 – 1,165.60

1779 votes

Oct 28, 2026

Category: creative_writing
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,150.36 rating

Confidence interval: 1,140.17 – 1,160.56

6411 votes

Oct 28, 2026

Category: english
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

966.30 rating

Confidence interval: 953.30 – 979.30

5678 votes

Oct 28, 2026

Category: exclude_ties
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,082.65 rating

Confidence interval: 1,038.13 – 1,127.16

193 votes

Oct 28, 2026

Category: german
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,079.03 rating

Confidence interval: 1,063.92 – 1,094.13

1752 votes

Oct 28, 2026

Category: hard_prompts
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,096.08 rating

Confidence interval: 1,079.80 – 1,112.35

1475 votes

Oct 28, 2026

Category: hard_prompts_english
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,078.41 rating

Confidence interval: 1,054.71 – 1,102.11

682 votes

Oct 28, 2026

Category: industry_business_and_management_and_financial_operations
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,112.23 rating

Confidence interval: 1,094.94 – 1,129.52

1539 votes

Oct 28, 2026

Category: industry_entertainment_and_sports_and_media
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,129.27 rating

Confidence interval: 1,100.14 – 1,158.40

434 votes

Oct 28, 2026

Category: industry_legal_and_government
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,109.33 rating

Confidence interval: 1,091.51 – 1,127.15

1334 votes

Oct 28, 2026

Category: industry_life_and_physical_and_social_science
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,107.36 rating

Confidence interval: 1,086.82 – 1,127.89

764 votes

Oct 28, 2026

Category: industry_mathematical
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,094.59 rating

Confidence interval: 1,063.99 – 1,125.19

385 votes

Oct 28, 2026

Category: industry_medicine_and_healthcare
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,088.34 rating

Confidence interval: 1,073.19 – 1,103.49

1935 votes

Oct 28, 2026

Category: industry_software_and_it_services
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,144.51 rating

Confidence interval: 1,129.84 – 1,159.17

2313 votes

Oct 28, 2026

Category: industry_writing_and_literature_and_language
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,092.91 rating

Confidence interval: 1,079.62 – 1,106.19

2680 votes

Oct 28, 2026

Category: instruction_following
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,096.77 rating

Confidence interval: 1,065.54 – 1,127.99

355 votes

Oct 28, 2026

Category: longer_query
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,115.98 rating

Confidence interval: 1,096.90 – 1,135.06

903 votes

Oct 28, 2026

Category: math
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,108.32 rating

Confidence interval: 1,088.98 – 1,127.66

1178 votes

Oct 28, 2026

Category: multi_turn
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,077.38 rating

Confidence interval: 1,061.78 – 1,092.97

1803 votes

Oct 28, 2026

Category: non_english
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,119.50 rating

Confidence interval: 1,110.17 – 1,128.82

8214 votes

Oct 28, 2026

Category: overall
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

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

WizardLM 70B

1,155.14 rating

Confidence interval: 1,110.49 – 1,199.79

172 votes

Oct 28, 2026

Category: russian
Evaluated configuration: WizardLM 70B
Evaluation type: human preference

Run it locally

Inference · memory estimate

WizardLM 70B

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license: Llama 2 Community