# Gemini 3.1 Pro Preview vs Gemini 3.5 Flash

> Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 54.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-1-pro-preview-vs-gemini-3-5-flash
- Last updated: 2026-10-10
- Shared benchmarks: 53

## Summary

- They share 53 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 7 categories and Gemini 3.5 Flash in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Gemini 3.1 Pro Preview leads 37.7 to 24.8.
- The biggest single-benchmark swing is DeepSWE: 11.7% for Gemini 3.1 Pro Preview and 37.4% for Gemini 3.5 Flash.
- Gemini 3.5 Flash is cheaper at $1.50 / $9 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.

## Snapshot

| | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| Provider | Google | Google |
| Noometry Index | 56.7 | 54.1 |
| Rank | 20 | 29 |
| Context | 1.05M | 1.05M |
| Input $/M | $2 | $1.50 |
| Output $/M | $12 | $9 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.1 Pro Preview: 42.7 (#88)
- Gemini 3.5 Flash: 49.1 (#46)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| SWE-bench Verified | 75.6% | 79.3% |
| DeepSWE | 11.7% | 37.4% |
| LMArena WebDev | 1446 | 1499 |
| SciCode | 58.9% | 53.1% |
| WeirdML | 72.1% | 62.6% |
| LMArena Coding | 1483 | 1492 |
| ALE-Bench | 1,161 | 911.02 |
| GSO | 22.6% | — |
| MirrorCode | 8.9% | — |
| AlgoTune | 2.02 | — |

## Agentic & Tool Use

- Gemini 3.1 Pro Preview: 37.7 (#33)
- Gemini 3.5 Flash: 24.8 (#107)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| APEX-Agents | 35.3% | 27.5% |
| GBAEval | 0.8% | 6.7% |
| GDP.pdf | 17% | 14% |
| Vending-Bench 2 | 3,774 | 5,396 |
| Terminal-Bench | 80.2% | — |
| τ²-bench Banking | 26% | — |
| DeepResearch Bench | 47.8% | — |
| PostTrainBench | 22% | — |
| BALROG | 57% | — |
| ExploitBench | 26.1% | — |
| LMArena Search | 1211 | — |
| METR Time Horizons | 77% | — |

## Reasoning

- Gemini 3.1 Pro Preview: 71.7 (#9)
- Gemini 3.5 Flash: 62.7 (#16)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| ARC-AGI-2 | 77.1% | 72.1% |
| SimpleBench | 79.6% | 76.7% |
| NYT Connections (extended) | 97.4% | 92.6% |
| ARC-AGI-1 | 98% | 92.5% |
| CritPt | 17.7% | 13.1% |
| Chess Puzzles | 55% | 50% |
| EnigmaEval | 36.8% | 25.4% |
| EBR-Bench | 14.3% | 4.8% |
| LMArena Hard Prompts | 1484 | 1488 |
| Mystery Game Puzzles | 34% | 32% |
| DTBench | 97.1% | 94.7% |
| LMCA | 53.8% | 47.1% |
| Epoch Capabilities Index | 154.92 | 154.55 |
| ForecastBench | 59 | 59 |
| Thematic Generalization | 79.4% | — |
| Surface Evolver Bench | — | 58.1% |

## Math

- Gemini 3.1 Pro Preview: 61.8 (#28)
- Gemini 3.5 Flash: 60.5 (#30)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.6% | 62.8% |
| FrontierMath Tier 4 | 26.8% | 26.8% |
| MathArena Final-Answer Competitions | 86.5% | 76.3% |
| OTIS Mock AIME 2024-2025 | 95.6% | 95.6% |
| ProofBench | 26% | 31% |
| LMArena Math | 1485 | 1502 |
| FrontierMath (Feb 2025 set) | 36.9% | 39% |
| FrontierMath Tier 4 (v1) | 16.7% | 14.6% |

## Knowledge

- Gemini 3.1 Pro Preview: 71.8 (#3)
- Gemini 3.5 Flash: 66.3 (#10)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| GPQA Diamond | 94.4% | 92.8% |
| SimpleQA Verified | 73.5% | 66.2% |
| LMArena Expert | 1484 | 1493 |
| Humanity's Last Exam | 46.4% | — |
| Vectara Hallucination Rate | 10.4% | — |

## Multimodal

- Gemini 3.1 Pro Preview: 38.1 (#60)
- Gemini 3.5 Flash: 45.7 (#13)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| LMArena Vision | 1296 | 1311 |
| Blueprint-Bench 2 | 26.5% | 33.6% |
| LMArena Document | 1444 | 1463 |
| Furniture Assembly | 26.7% | — |

## Multilingual

- Gemini 3.1 Pro Preview: 57.1 (#11)
- Gemini 3.5 Flash: 57.1 (#12)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1475 | 1475 |
| LMArena Chinese | 1528 | 1525 |
| LMArena French | 1488 | 1490 |
| LMArena German | 1491 | 1500 |
| LMArena Japanese | 1494 | 1481 |
| LMArena Korean | 1458 | 1453 |
| LMArena Russian | 1496 | 1491 |
| LMArena Spanish | 1480 | 1478 |

## Instruction Following

- Gemini 3.1 Pro Preview: 77.1 (#31)
- Gemini 3.5 Flash: 77.2 (#29)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1465 | 1466 |

## Long Context

- Gemini 3.1 Pro Preview: 47.3 (#17)
- Gemini 3.5 Flash: 45.3 (#39)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1482 | 1481 |
| CL-bench | 20.8% | — |
| CL-bench Life | 16.9% | — |

## Writing & Preference

- Gemini 3.1 Pro Preview: 66.3 (#36)
- Gemini 3.5 Flash: 65.5 (#46)

| Benchmark | Gemini 3.1 Pro Preview | Gemini 3.5 Flash |
|---|---|---|
| LMArena Text | 1480 | 1481 |
| LMArena Creative Writing | 1482 | 1467 |
| EQ-Bench 4 | 1142 | 1087 |
| LMArena Multi-Turn | 1486 | 1482 |
| EQ-Bench Creative Writing | 1491 | — |

## FAQ

### Is Gemini 3.1 Pro Preview better than Gemini 3.5 Flash?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 54.1 on the Noometry Index.

### Which is cheaper, Gemini 3.1 Pro Preview or Gemini 3.5 Flash?

Gemini 3.5 Flash is cheaper. It lists at $1.50 per million input tokens and $9 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

### Is Gemini 3.1 Pro Preview or Gemini 3.5 Flash better for coding?

Gemini 3.5 Flash scores higher on coding benchmarks: 49.1 versus 42.7 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

### How many benchmarks do Gemini 3.1 Pro Preview and Gemini 3.5 Flash share?

53 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and Gemini 3.5 Flash has 54.