# Gemini 3.1 Pro Preview vs Step 3.7 Flash

> Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 36.5 on the Noometry Index. Step 3.7 Flash costs 11× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-3-1-pro-preview-vs-step-3-7-flash
- Last updated: 2026-10-10
- Shared benchmarks: 5

## Summary

- They share 5 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 3 categories and Step 3.7 Flash in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 20.9.
- The biggest single-benchmark swing is NYT Connections (extended): 97.4% for Gemini 3.1 Pro Preview and 39.7% for Step 3.7 Flash.
- Step 3.7 Flash is cheaper at $0.18 / $1.11 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
- Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 256K.
- Step 3.7 Flash has downloadable open weights; the other is API-only.

## Snapshot

| | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| Provider | Google | StepFun |
| Noometry Index | 56.7 | 36.5 |
| Rank | 20 | 207 |
| Context | 1.05M | 256K |
| Input $/M | $2 | $0.18 |
| Output $/M | $12 | $1.11 |
| Weights | Proprietary | Open |

## Coding

- Gemini 3.1 Pro Preview: 42.7 (#88)
- Step 3.7 Flash: 39.4 (#147)

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| SciCode | 58.9% | 40% |
| ALE-Bench | 1,161 | 694.12 |
| SWE-bench Verified | 75.6% | — |
| DeepSWE | 11.7% | — |
| LMArena WebDev | 1446 | — |
| GSO | 22.6% | — |
| WeirdML | 72.1% | — |
| LMArena Coding | 1483 | — |
| MirrorCode | 8.9% | — |
| AlgoTune | 2.02 | — |

## Agentic & Tool Use

- Gemini 3.1 Pro Preview: 37.7 (#33)
- Step 3.7 Flash: —

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| Terminal-Bench | 80.2% | — |
| APEX-Agents | 35.3% | — |
| τ²-bench Banking | 26% | — |
| DeepResearch Bench | 47.8% | — |
| PostTrainBench | 22% | — |
| BALROG | 57% | — |
| ExploitBench | 26.1% | — |
| GBAEval | 0.8% | — |
| GDP.pdf | 17% | — |
| LMArena Search | 1211 | — |
| METR Time Horizons | 77% | — |
| Vending-Bench 2 | 3,774 | — |

## Reasoning

- Gemini 3.1 Pro Preview: 71.7 (#9)
- Step 3.7 Flash: 20.9 (#211)

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| NYT Connections (extended) | 97.4% | 39.7% |
| CritPt | 17.7% | 2.3% |
| ARC-AGI-2 | 77.1% | — |
| SimpleBench | 79.6% | — |
| ARC-AGI-1 | 98% | — |
| Chess Puzzles | 55% | — |
| EnigmaEval | 36.8% | — |
| Thematic Generalization | 79.4% | — |
| EBR-Bench | 14.3% | — |
| LMArena Hard Prompts | 1484 | — |
| Mystery Game Puzzles | 34% | — |
| DTBench | 97.1% | — |
| LMCA | 53.8% | — |
| Epoch Capabilities Index | 154.92 | — |
| ForecastBench | 59 | — |

## Math

- Gemini 3.1 Pro Preview: 61.8 (#28)
- Step 3.7 Flash: 40.9 (#74)

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| MathArena Final-Answer Competitions | 86.5% | 68.5% |
| FrontierMath (Tiers 1-3) | 59.6% | — |
| FrontierMath Tier 4 | 26.8% | — |
| OTIS Mock AIME 2024-2025 | 95.6% | — |
| ProofBench | 26% | — |
| LMArena Math | 1485 | — |
| FrontierMath (Feb 2025 set) | 36.9% | — |
| FrontierMath Tier 4 (v1) | 16.7% | — |

## Knowledge

- Gemini 3.1 Pro Preview: 71.8 (#3)
- Step 3.7 Flash: —

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| GPQA Diamond | 94.4% | — |
| Humanity's Last Exam | 46.4% | — |
| SimpleQA Verified | 73.5% | — |
| Vectara Hallucination Rate | 10.4% | — |
| LMArena Expert | 1484 | — |

## Multimodal

- Gemini 3.1 Pro Preview: 38.1 (#60)
- Step 3.7 Flash: —

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| LMArena Vision | 1296 | — |
| Blueprint-Bench 2 | 26.5% | — |
| Furniture Assembly | 26.7% | — |
| LMArena Document | 1444 | — |

## Multilingual

- Gemini 3.1 Pro Preview: 57.1 (#11)
- Step 3.7 Flash: —

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1475 | — |
| LMArena Chinese | 1528 | — |
| LMArena French | 1488 | — |
| LMArena German | 1491 | — |
| LMArena Japanese | 1494 | — |
| LMArena Korean | 1458 | — |
| LMArena Russian | 1496 | — |
| LMArena Spanish | 1480 | — |

## Instruction Following

- Gemini 3.1 Pro Preview: 77.1 (#31)
- Step 3.7 Flash: —

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1465 | — |

## Long Context

- Gemini 3.1 Pro Preview: 47.3 (#17)
- Step 3.7 Flash: —

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| CL-bench | 20.8% | — |
| CL-bench Life | 16.9% | — |
| LMArena Longer Query | 1482 | — |

## Writing & Preference

- Gemini 3.1 Pro Preview: 66.3 (#36)
- Step 3.7 Flash: —

| Benchmark | Gemini 3.1 Pro Preview | Step 3.7 Flash |
|---|---|---|
| LMArena Text | 1480 | — |
| LMArena Creative Writing | 1482 | — |
| EQ-Bench Creative Writing | 1491 | — |
| EQ-Bench 4 | 1142 | — |
| LMArena Multi-Turn | 1486 | — |

## FAQ

### Is Gemini 3.1 Pro Preview better than Step 3.7 Flash?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 36.5 on the Noometry Index. Step 3.7 Flash costs 11× less per token, which makes it the better buy when Gemini 3.1 Pro Preview's lead doesn't matter for your workload.

### Which is cheaper, Gemini 3.1 Pro Preview or Step 3.7 Flash?

Step 3.7 Flash is cheaper. It lists at $0.18 per million input tokens and $1.11 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

### Is Gemini 3.1 Pro Preview or Step 3.7 Flash better for coding?

Gemini 3.1 Pro Preview scores higher on coding benchmarks: 42.7 versus 39.4 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 256K.

### How many benchmarks do Gemini 3.1 Pro Preview and Step 3.7 Flash share?

5 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Step 3.7 Flash has 5.