StepFun: Step 5 Preview vs Sakana: Fugu Max
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-10-09.
| StepFun: Step 5 Preview | Sakana: Fugu Max | |
|---|---|---|
| Vendor | stepfun | sakana |
| Quality Score | 100 | 100 |
| Benchmark Score | 74.2 | - |
| Input Price | $1.00/M | $2.00/M |
| Output Price | $2.70/M | $6.00/M |
| Context Window | 1,000,000 | 1,000,000 |
| Max Output | 64,000 | 128,000 |
| Tool Calling | ✓ | ✓ |
| Structured Output | ✓ | ✓ |
| Reasoning Mode | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio | - | - |
| Benchmark Scores | ||
| ai_index | 72.2 | - |
The verdict
For a sample job of 1 million input tokens and 250,000 output tokens, Step 5 Preview costs $1.68 and Fugu Max costs $3.50, so Step 5 Preview is about 2.1 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 20 tasks where they differ, Step 5 Preview scores higher on 20 (SQL Generation, Code Review, Code Completion and others).
Step 5 Preview was released in October 2026 and Fugu Max in September 2026.
Pick Step 5 Preview if you need lower cost, video input or the stronger all-round task scores.
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?
| Task | StepFun: Step 5 Preview | Sakana: Fugu Max |
|---|---|---|
| SQL Generation | 142 | 133 |
| Code Review | 145 | 132 |
| Code Completion | 120 | 118 |
| Code Refactoring | 147 | 136 |
| Bug Fixing | 149 | 136 |
| Unit Test Generation | 132 | 124 |
| Code Documentation | 137 | 130 |
| Regex Writing | 126 | 118 |
| CI/CD Pipelines | 128 | 120 |
| Frontend Component Design | 130 | 122 |
| Data Analysis | 134 | 124 |
| CSV / Spreadsheet Cleanup | 136 | 133 |
| ETL Scripting | 138 | 128 |
| JSON Extraction | 123 | 122 |
| Bulk Data Labeling | 120 | 119 |
| OCR / Document Parsing | 134 | 131 |
| Table Extraction from PDFs | 134 | 131 |
| Long-Document Summarization | 149 | 137 |
| Short-Form Summarization | 118 | 114 |
| Blog Post Writing | 130 | 121 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →