head-to-head

StepFun: Step 5 Preview vs Anthropic: Claude Haiku 5.5 (batch)

Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-10-09.

StepFun: Step 5 Preview Anthropic: Claude Haiku 5.5 (batch)
Vendorstepfunanthropic
Quality Score100100
Benchmark Score74.273.6
Input Price$1.00/M$0.05/M
Output Price$2.70/M$0.25/M
Context Window1,000,0001,000,000
Max Output64,000128,000
Tool Calling✓✓
Structured Output✓✓
Reasoning Mode✓✓
Vision✓✓
Audio--
Benchmark Scores
ai_index72.271.6

The verdict

For a sample job of 1 million input tokens and 250,000 output tokens, Claude Haiku 5.5 (batch) costs $0.11 and Step 5 Preview costs $1.68, so Claude Haiku 5.5 (batch) is about 14.9 times cheaper for the same work.

Both read up to 1,000,000 tokens at once, roughly 1,500 pages of text.

Only Step 5 Preview offers video input.

Across the 18 tasks where they differ, Step 5 Preview scores higher on 8 (Code Review, Code Refactoring, Bug Fixing and others), Claude Haiku 5.5 (batch) on 10 (SQL Generation, Code Completion, Code Documentation and others).

Pick Step 5 Preview if you need video input; pick Claude Haiku 5.5 (batch) if you need lower cost.

Costs use current list prices per million tokens; page counts assume about 0.75 words per token and 500 words per page.

Who wins by task?

TaskStepFun: Step 5 PreviewAnthropic: Claude Haiku 5.5 (batch)
SQL Generation 142 142
Code Review 145 145
Code Completion 120 133
Code Refactoring 147 147
Bug Fixing 149 149
Unit Test Generation 132 132
Code Documentation 137 138
Regex Writing 126 127
CI/CD Pipelines 128 128
Frontend Component Design 130 130
Data Analysis 134 134
CSV / Spreadsheet Cleanup 136 136
ETL Scripting 138 138
JSON Extraction 123 131
Bulk Data Labeling 120 129
Long-Document Summarization 149 149
Short-Form Summarization 118 127
Blog Post Writing 130 130

Scores reflect capability match + benchmark data + pricing for each task. Methodology →

Related comparisons

StepFun: Step 5 Preview vs Anthropic: Claude Haiku 5.5 StepFun: Step 5 Preview vs Mistral: Mistral Large 4 StepFun: Step 5 Preview vs OpenAI: GPT-6.1 Sol Pro StepFun: Step 5 Preview vs OpenAI: GPT-6.1 Sol Pro (batch) StepFun: Step 5 Preview vs OpenAI: GPT-6.1 Sol StepFun: Step 5 Preview vs OpenAI: GPT-6.1 Sol (batch) StepFun: Step 5 Preview vs Anthropic: Claude Sonnet 5.5 StepFun: Step 5 Preview vs Anthropic: Claude Sonnet 5.5 (batch)

The Model Movers Report

One email every Friday, built from this site's own rankings: the current top five by benchmark score, every model released in the last seven days, and one note worked out from that week's numbers. You can unsubscribe from any issue with one click.