GeekRadar / The comparison desk
COMPARE.
AI models
stepfun

step-2-16k-exp-202412

Model evaluation

Capability & preference

Human preference · Arena

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

step-2-16k-exp-202412 1320.9 step-2-16k-exp-202412 · 95% interval 1312.3–1329.4 · 4,833 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

step-2-16k-exp-202412

1,331.19 rating

Confidence interval: 1,298.98 – 1,363.40

328 votes

Category
chinese
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,316.02 rating

Confidence interval: 1,296.36 – 1,335.68

737 votes

Category
coding
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,327.45 rating

Confidence interval: 1,306.83 – 1,348.08

751 votes

Category
creative_writing
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,344.36 rating

Confidence interval: 1,333.60 – 1,355.12

2781 votes

Category
english
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,271.89 rating

Confidence interval: 1,258.18 – 1,285.61

3048 votes

Category
exclude_ties
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,278.41 rating

Confidence interval: 1,247.62 – 1,309.21

310 votes

Category
expert
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,298.95 rating

Confidence interval: 1,283.73 – 1,314.17

1221 votes

Category
hard_prompts
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,308.36 rating

Confidence interval: 1,289.84 – 1,326.87

786 votes

Category
hard_prompts_english
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,283.05 rating

Confidence interval: 1,258.43 – 1,307.67

506 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,291.36 rating

Confidence interval: 1,273.09 – 1,309.63

873 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,336.35 rating

Confidence interval: 1,303.01 – 1,369.69

278 votes

Category
industry_legal_and_government
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,329.88 rating

Confidence interval: 1,310.49 – 1,349.27

863 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,310.13 rating

Confidence interval: 1,287.84 – 1,332.43

543 votes

Category
industry_mathematical
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,354.79 rating

Confidence interval: 1,319.63 – 1,389.95

265 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,315.09 rating

Confidence interval: 1,299.46 – 1,330.72

1200 votes

Category
industry_software_and_it_services
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,303.48 rating

Confidence interval: 1,288.03 – 1,318.93

1326 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,286.62 rating

Confidence interval: 1,274.08 – 1,299.16

1950 votes

Category
instruction_following
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,306.27 rating

Confidence interval: 1,285.22 – 1,327.32

693 votes

Category
longer_query
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,304.42 rating

Confidence interval: 1,284.54 – 1,324.29

642 votes

Category
math
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,292.34 rating

Confidence interval: 1,272.33 – 1,312.36

805 votes

Category
multi_turn
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,289.96 rating

Confidence interval: 1,277.40 – 1,302.52

2052 votes

Category
non_english
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,320.87 rating

Confidence interval: 1,312.33 – 1,329.41

4833 votes

Category
overall
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

step-2-16k-exp-202412

1,323.95 rating

Confidence interval: 1,302.23 – 1,345.67

642 votes

Category
russian
Evaluated configuration
step-2-16k-exp-202412
Evaluation type
human preference

Run it locally

Inference · memory estimate

step-2-16k-exp-202412

Not enough verified architecture information for a reliable VRAM estimate.

Capabilities

Model license
Proprietary