# Gemini 2.5 Flash vs Gemini 3.1 Pro Preview

> Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 5.3× 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-2-5-flash-vs-gemini-3-1-pro-preview
- Last updated: 2026-10-10
- Shared benchmarks: 38

## Summary

- They share 38 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 2 categories and Gemini 3.1 Pro Preview in 8 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 18.1.
- The biggest single-benchmark swing is ARC-AGI-2: 2.5% for Gemini 2.5 Flash and 77.1% for Gemini 3.1 Pro Preview.
- Gemini 2.5 Flash is cheaper at $0.30 / $2.50 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.

## Snapshot

| | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| Provider | Google | Google |
| Noometry Index | 39.3 | 56.7 |
| Rank | 170 | 23 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.30 | $2 |
| Output $/M | $2.50 | $12 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 2.5 Flash: 35.8 (#220)
- Gemini 3.1 Pro Preview: 42.5 (#99)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| WeirdML | 41.9% | 72.1% |
| LMArena Coding | 1424 | 1484 |
| ALE-Bench | 661.88 | 1,161 |
| SWE-bench Verified | — | 75.6% |
| DeepSWE | — | 11.7% |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| LMArena WebDev | — | 1447 |
| SciCode | — | 58.9% |
| GSO | — | 22.6% |
| MirrorCode | — | 8.9% |
| AlgoTune | — | 2.02 |

## Agentic & Tool Use

- Gemini 2.5 Flash: 30.8 (#74)
- Gemini 3.1 Pro Preview: 37.7 (#34)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| Terminal-Bench | 17.1% | 80.2% |
| BALROG | 33.5% | 57% |
| Vending-Bench 2 | 548.84 | 3,774 |
| APEX-Agents | — | 35.3% |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| τ²-bench Banking | — | 26% |
| DeepResearch Bench | — | 47.8% |
| PostTrainBench | — | 22% |
| ExploitBench | — | 26.1% |
| GBAEval | — | 0.8% |
| GDP.pdf | — | 17% |
| LMArena Search | — | 1211 |
| METR Time Horizons | — | 77% |

## Reasoning

- Gemini 2.5 Flash: 18.1 (#286)
- Gemini 3.1 Pro Preview: 71.7 (#12)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| ARC-AGI-2 | 2.5% | 77.1% |
| SimpleBench | 41.2% | 79.6% |
| ARC-AGI-1 | 33.3% | 98% |
| CritPt | 1.1% | 17.7% |
| EnigmaEval | 2.7% | 36.8% |
| LMArena Hard Prompts | 1422 | 1485 |
| DTBench | 76.5% | 97.1% |
| LMCA | 27.5% | 53.8% |
| Epoch Capabilities Index | 143.03 | 154.77 |
| ForecastBench | 60.6 | 59 |
| Kagi LLM Benchmark | 56.8% | — |
| NYT Connections (extended) | — | 97.4% |
| Chess Puzzles | — | 55% |
| Thematic Generalization | — | 79.4% |
| EBR-Bench | — | 14.3% |
| Mystery Game Puzzles | — | 34% |

## Math

- Gemini 2.5 Flash: 39.9 (#98)
- Gemini 3.1 Pro Preview: 62.1 (#34)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 95.6% |
| LMArena Math | 1415 | 1485 |
| FrontierMath (Feb 2025 set) | 4.8% | 36.9% |
| FrontierMath Tier 4 (v1) | 4.2% | 16.7% |
| FrontierMath (Tiers 1-3) | — | 59.6% |
| FrontierMath Tier 4 | — | 26.8% |
| MathArena Final-Answer Competitions | — | 86.5% |
| ProofBench | — | 26% |
| Omni-MATH | 38.5% | — |

## Knowledge

- Gemini 2.5 Flash: 36.4 (#168)
- Gemini 3.1 Pro Preview: 71.8 (#3)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| Humanity's Last Exam | 12.1% | 46.4% |
| Vectara Hallucination Rate | 7.8% | 10.4% |
| LMArena Expert | 1426 | 1485 |
| GPQA Diamond | — | 94.4% |
| SimpleQA Verified | — | 73.5% |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| GPQA (HELM) | 39% | — |

## Multimodal

- Gemini 2.5 Flash: 41.8 (#32)
- Gemini 3.1 Pro Preview: 37.9 (#69)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Vision | 1253 | 1296 |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| Blueprint-Bench 2 | — | 26.5% |
| Furniture Assembly | — | 26.7% |
| LMArena Document | — | 1444 |
| SpatialViz-Bench | 36.9% | — |

## Multilingual

- Gemini 2.5 Flash: 52.3 (#88)
- Gemini 3.1 Pro Preview: 57.0 (#12)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Non-English | 1409 | 1477 |
| LMArena Chinese | 1450 | 1529 |
| LMArena French | 1433 | 1487 |
| LMArena German | 1418 | 1491 |
| LMArena Japanese | 1405 | 1493 |
| LMArena Korean | 1385 | 1455 |
| LMArena Russian | 1415 | 1498 |
| LMArena Spanish | 1421 | 1479 |

## Instruction Following

- Gemini 2.5 Flash: 75.7 (#54)
- Gemini 3.1 Pro Preview: 77.0 (#32)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Instruction Following | 1405 | 1466 |
| IFEval | 89.8% | — |

## Long Context

- Gemini 2.5 Flash: 47.5 (#17)
- Gemini 3.1 Pro Preview: 47.4 (#18)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Longer Query | 1419 | 1483 |
| Fiction.LiveBench | 77.8% | — |
| CL-bench | — | 20.8% |
| CL-bench Life | — | 16.9% |

## Writing & Preference

- Gemini 2.5 Flash: 53.8 (#157)
- Gemini 3.1 Pro Preview: 66.1 (#37)

| Benchmark | Gemini 2.5 Flash | Gemini 3.1 Pro Preview |
|---|---|---|
| LMArena Text | 1417 | 1481 |
| LMArena Creative Writing | 1400 | 1482 |
| EQ-Bench Creative Writing | 1137 | 1491 |
| LMArena Multi-Turn | 1408 | 1488 |
| Short-Story Creative Writing | 76.5% | — |
| WildBench | 81.7% | — |
| EQ-Bench 4 | — | 1142 |

## FAQ

### Is Gemini 2.5 Flash better than Gemini 3.1 Pro Preview?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 39.3 on the Noometry Index. Gemini 2.5 Flash costs 5.3× 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 2.5 Flash or Gemini 3.1 Pro Preview?

Gemini 2.5 Flash is cheaper. It lists at $0.30 per million input tokens and $2.50 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

### Is Gemini 2.5 Flash or Gemini 3.1 Pro Preview better for coding?

Gemini 3.1 Pro Preview scores higher on coding benchmarks: 42.5 versus 35.8 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

### How many benchmarks do Gemini 2.5 Flash and Gemini 3.1 Pro Preview share?

38 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Gemini 3.1 Pro Preview has 71.