GeekRadar / The comparison desk
COMPARE.
AI models
stepfun

Step-1o Turbo (202506)

Model evaluation

Capability & preference

Human preference · Arena

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

Step-1o Turbo (202506) 1335.6 Step-1o Turbo (202506) · 95% interval 1328.8–1342.5 · 8,902 votes · 2026-11-08

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 · 2026-11-08 · rating

Step-1o Turbo (202506)

1,379.65 rating

Confidence interval: 1,354.49 – 1,404.80

546 votes

Category
chinese
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,339.24 rating

Confidence interval: 1,324.43 – 1,354.06

1493 votes

Category
coding
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,306.95 rating

Confidence interval: 1,288.09 – 1,325.81

993 votes

Category
creative_writing
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,352.77 rating

Confidence interval: 1,343.11 – 1,362.44

3990 votes

Category
english
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,288.23 rating

Confidence interval: 1,278.75 – 1,297.71

6356 votes

Category
exclude_ties
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,307.25 rating

Confidence interval: 1,280.71 – 1,333.79

483 votes

Category
expert
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,314.18 rating

Confidence interval: 1,282.76 – 1,345.60

345 votes

Category
german
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,334.98 rating

Confidence interval: 1,325.14 – 1,344.82

3696 votes

Category
hard_prompts
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,354.24 rating

Confidence interval: 1,340.22 – 1,368.26

1704 votes

Category
hard_prompts_english
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,328.63 rating

Confidence interval: 1,312.92 – 1,344.35

1384 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,292.95 rating

Confidence interval: 1,277.98 – 1,307.93

1469 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,354.66 rating

Confidence interval: 1,331.53 – 1,377.78

654 votes

Category
industry_legal_and_government
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,358.15 rating

Confidence interval: 1,342.96 – 1,373.34

1506 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,344.64 rating

Confidence interval: 1,320.43 – 1,368.85

568 votes

Category
industry_mathematical
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,349.14 rating

Confidence interval: 1,324.91 – 1,373.37

598 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,353.41 rating

Confidence interval: 1,342.28 – 1,364.54

2789 votes

Category
industry_software_and_it_services
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,309.17 rating

Confidence interval: 1,295.48 – 1,322.86

1830 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,309.54 rating

Confidence interval: 1,296.78 – 1,322.29

2096 votes

Category
instruction_following
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,348.28 rating

Confidence interval: 1,331.98 – 1,364.58

1272 votes

Category
longer_query
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,317.88 rating

Confidence interval: 1,294.15 – 1,341.60

564 votes

Category
math
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,340.19 rating

Confidence interval: 1,324.05 – 1,356.34

1356 votes

Category
multi_turn
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,312.52 rating

Confidence interval: 1,303.71 – 1,321.33

4912 votes

Category
non_english
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,335.62 rating

Confidence interval: 1,328.76 – 1,342.47

8902 votes

Category
overall
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,303.64 rating

Confidence interval: 1,288.13 – 1,319.15

1583 votes

Category
polish
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-11-08 · rating

Step-1o Turbo (202506)

1,327.00 rating

Confidence interval: 1,305.83 – 1,348.16

653 votes

Category
russian
Evaluated configuration
Step-1o Turbo (202506)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Step-1o Turbo (202506)

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary