Gemini 3.7 Flash vs Step 5 Preview
Gemini 3.7 Flash 
Step 5 Preview Verdict
- Gemini 3.7 Flash (medium) and Step 5 Preview are level on quality (64.7 vs 65.0).
- Gemini 3.7 Flash (medium) is stronger in agents & tools, coding, knowledge, multimodal.
- Step 5 Preview is stronger in composite, long context, reasoning.
- They cost about the same ($1.43 per 1M blended).
| Metric | Gemini 3.7 Flash (medium) | Step 5 Preview |
|---|---|---|
| BenchLeader Index | 64.7 | 65.0 |
| Agents & tools score | 64.1 | – |
| Coding score | 68.5 | – |
| Composite score | 79.2 | 84.4 |
| Knowledge score | 75.5 | 72.0 |
| Long context score | 68.0 | 70.8 |
| Multimodal score | 69.4 | 60.8 |
| Reasoning score | 62.1 | 82.1 |
| Blended price $/M | $1.50 | $1.43 |
| Output speed | 276 tok/s | – |
| Time to first answer | 5.4 s | – |
| Context window | 1.0M | 1M |
| SciCode | 57.9% | – |
| AA Intelligence Index | 39.6 | 43.6 |
| AA-LCR | 83.0% | 88.3% |
| MMMU-Pro | 84.7% | 76.4% |
| AA-Omniscience | 23.7 | 16.4 |
| GPQA Diamond (AA) | 92.1% | – |
| Humanity's Last Exam (AA) | 39.0% | 46.5% |
| SciCode (AA) | 59.8% | 58.9% |
| ARC-AGI-1 | 91.2% | – |
| ARC-AGI-2 | 63.8% | – |
| CritPt | 9.4% | 20.9% |
| GDPval (AA) | 45.3% | 53.6% |
| τ²-Bench Banking (AA) | 35.5% | – |
| DeepSWE | 65.5% | – |
| CursorBench | 59.0% | – |
| GDP.pdf | 21.8% | – |
Data as of 2026-09-19. Best configuration of each model; every score links to its source on the model pages.
Gemini 3.7 Flash vs Step 5 Preview: questions
- Is Gemini 3.7 Flash better than Step 5 Preview?
- Gemini 3.7 Flash (medium) and Step 5 Preview are level on quality (64.7 vs 65.0). The BenchLeader Index combines every independent quality benchmark; Step 5 Preview is ahead overall as of 2026-09-19, but check the category scores for your use.
- Which is cheaper, Gemini 3.7 Flash or Step 5 Preview?
- Step 5 Preview is cheaper: $1.43 against $1.50 per million tokens, blended at three input tokens per output token.
- Which has the larger context window?
- Gemini 3.7 Flash accepts more context: 1.0M against 1M tokens.