StepFun: Step 3.7 Flash vs Google: Gemini 3.5 Flash
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-11-07.
| StepFun: Step 3.7 Flash | Google: Gemini 3.5 Flash | |
|---|---|---|
| Vendor | stepfun | |
| Quality Score | 100 | 100 |
| Benchmark Score | 50.8 | 84.2 |
| Input Price | $0.20/M | $1.50/M |
| Output Price | $1.15/M | $9.00/M |
| Context Window | 262,144 | 1,048,576 |
| Max Output | 230,400 | 65,536 |
| Tool Calling | ✓ | ✓ |
| Structured Output | ✓ | ✓ |
| Reasoning Mode | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio | - | ✓ |
| Benchmark Scores | ||
| ai_index | 49.9 | 82.8 |
| ai_index_agentic | 35.5 | 61.8 |
| ai_index_coding | 65.3 | 100.0 |
The verdict
For a sample job of 1 million input tokens and 250,000 output tokens, Step 3.7 Flash costs $0.49 and Gemini 3.5 Flash costs $3.75, so Step 3.7 Flash is about 7.7 times cheaper for the same work.
Gemini 3.5 Flash reads up to 1,048,576 tokens at once, roughly 1,573 pages of text, against about 393 pages for Step 3.7 Flash.
Only Gemini 3.5 Flash offers audio input.
On our blended benchmark score Gemini 3.5 Flash leads 84.2 to 50.8.
Across the 20 tasks where they differ, Step 3.7 Flash scores higher on 4 (Code Completion, JSON Extraction, Bulk Data Labeling and others), Gemini 3.5 Flash on 16 (SQL Generation, Code Review, Code Refactoring and others).
Pick Step 3.7 Flash if you need lower cost; pick Gemini 3.5 Flash if you need long documents, audio input, benchmark performance 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 3.7 Flash | Google: Gemini 3.5 Flash |
|---|---|---|
| SQL Generation | 153 | 168 |
| Code Review | 146 | 164 |
| Code Completion | 130 | 119 |
| Code Refactoring | 144 | 162 |
| Bug Fixing | 155 | 176 |
| Unit Test Generation | 139 | 152 |
| Code Documentation | 133 | 141 |
| Regex Writing | 129 | 133 |
| CI/CD Pipelines | 131 | 143 |
| Frontend Component Design | 136 | 145 |
| Data Analysis | 150 | 167 |
| CSV / Spreadsheet Cleanup | 141 | 153 |
| ETL Scripting | 137 | 152 |
| JSON Extraction | 142 | 138 |
| Bulk Data Labeling | 133 | 124 |
| OCR / Document Parsing | 138 | 146 |
| Table Extraction from PDFs | 138 | 146 |
| Long-Document Summarization | 142 | 157 |
| Short-Form Summarization | 128 | 121 |
| Blog Post Writing | 129 | 138 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →