# Gemini 2.5 Pro vs Gemini 3 Flash Preview

> Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 45.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-2-5-pro-vs-gemini-3-flash-preview
- Last updated: 2026-10-10
- Shared benchmarks: 48

## Summary

- They share 48 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 1 category and Gemini 3 Flash Preview in 9 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Flash Preview leads 49.2 to 28.8.
- The biggest single-benchmark swing is ARC-AGI-1: 41% for Gemini 2.5 Pro and 84.7% for Gemini 3 Flash Preview.
- Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $1.25 / $10 for Gemini 2.5 Pro.

## Snapshot

| | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| Provider | Google | Google |
| Noometry Index | 45.0 | 52.3 |
| Rank | 75 | 40 |
| Context | 1.05M | 1.05M |
| Input $/M | $1.25 | $0.50 |
| Output $/M | $10 | $3 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 2.5 Pro: 42.4 (#101)
- Gemini 3 Flash Preview: 50.9 (#42)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| SWE-bench Verified | 57.6% | 75.4% |
| SWE-bench Verified (bash only) | 53.6% | 75.8% |
| LMArena WebDev | 1227 | 1439 |
| GSO | 3.9% | 9.8% |
| WeirdML | 54% | 61.6% |
| LMArena Coding | 1452 | 1460 |
| ALE-Bench | 785.52 | 1,367 |
| Aider Polyglot | 83.1% | — |
| SWE-bench Multilingual | — | 72.7% |
| SciCode | 42.8% | — |
| LiveBench Coding | 85.9% | — |
| CadEval | 64% | — |
| AlgoTune | 1.51 | — |

## Agentic & Tool Use

- Gemini 2.5 Pro: 29.2 (#88)
- Gemini 3 Flash Preview: 38.7 (#29)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| Terminal-Bench | 32.6% | 64.3% |
| τ²-bench Banking | 13.7% | 27.3% |
| DeepResearch Bench | 42.8% | 49.8% |
| BALROG | 43.3% | 48.1% |
| LMArena Search | 1142 | 1198 |
| Vending-Bench 2 | 573.64 | 3,635 |
| GDPval | 23.3% | — |
| Remote Labor Index | 0.8% | — |
| TheAgentCompany | 30.3% | — |
| τ²-bench Airline | — | 82.5% |
| τ²-bench Retail | — | 76.8% |
| τ²-bench Telecom | — | 91.2% |
| GDP.pdf | — | 10% |
| METR Time Horizons | 55.4% | — |

## Reasoning

- Gemini 2.5 Pro: 28.8 (#99)
- Gemini 3 Flash Preview: 49.2 (#37)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| ARC-AGI-2 | 4.9% | 33.6% |
| SimpleBench | 62.4% | 61.1% |
| ARC-AGI-1 | 41% | 84.7% |
| Chess Puzzles | 20% | 40% |
| LMArena Hard Prompts | 1455 | 1465 |
| DTBench | 82.4% | 89.1% |
| LMCA | 34.8% | 43.1% |
| Epoch Capabilities Index | 145.32 | 151.8 |
| ForecastBench | 61.3 | 58.5 |
| Kagi LLM Benchmark | 70.3% | — |
| NYT Connections (extended) | — | 83.1% |
| CritPt | 2% | — |
| EnigmaEval | 5.6% | — |
| LiveBench Reasoning | 89.8% | — |
| Mystery Game Puzzles | — | 26% |
| LiveBench Data Analysis | 79.9% | — |
| LiveBench | 82.3% | — |

## Math

- Gemini 2.5 Pro: 32.5 (#213)
- Gemini 3 Flash Preview: 51.7 (#55)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| FrontierMath (Tiers 1-3) | 24.6% | 51.2% |
| FrontierMath Tier 4 | 0% | 17.1% |
| OTIS Mock AIME 2024-2025 | 84.7% | 95.6% |
| LMArena Math | 1450 | 1473 |
| FrontierMath (Feb 2025 set) | 14.1% | 35.6% |
| FrontierMath Tier 4 (v1) | 4.2% | 4.2% |
| MathArena Final-Answer Competitions | — | 67.6% |
| ProofBench | — | 15% |
| Omni-MATH | 41.6% | — |
| LiveBench Math | 90.2% | — |
| MATH Level 5 | 95.9% | — |

## Knowledge

- Gemini 2.5 Pro: 56.0 (#46)
- Gemini 3 Flash Preview: 58.8 (#33)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| GPQA Diamond | 85.3% | 89.4% |
| Vectara Hallucination Rate | 7% | 13.5% |
| LMArena Expert | 1452 | 1462 |
| Humanity's Last Exam | 21.6% | — |
| SimpleQA Verified | — | 66.8% |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.6% | — |
| GPQA (HELM) | 74.9% | — |

## Multimodal

- Gemini 2.5 Pro: 45.2 (#18)
- Gemini 3 Flash Preview: 45.5 (#16)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Vision | 1263 | 1285 |
| GeoBench | 86% | 88% |
| VPCT | 48% | 72.6% |
| LMArena Document | 1421 | 1413 |
| Blueprint-Bench 2 | — | 0% |
| SpatialViz-Bench | 44.7% | — |

## Multilingual

- Gemini 2.5 Pro: 55.3 (#31)
- Gemini 3 Flash Preview: 55.7 (#27)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Non-English | 1451 | 1458 |
| LMArena Chinese | 1507 | 1511 |
| LMArena French | 1472 | 1477 |
| LMArena German | 1487 | 1497 |
| LMArena Japanese | 1461 | 1489 |
| LMArena Korean | 1434 | 1443 |
| LMArena Russian | 1461 | 1480 |
| LMArena Spanish | 1473 | 1469 |

## Instruction Following

- Gemini 2.5 Pro: 75.0 (#75)
- Gemini 3 Flash Preview: 75.7 (#56)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Instruction Following | 1437 | 1437 |
| LiveBench Instruction Following | 80.6% | — |
| IFEval | 84% | — |

## Long Context

- Gemini 2.5 Pro: 59.8 (#5)
- Gemini 3 Flash Preview: 44.4 (#67)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Longer Query | 1449 | 1452 |
| Fiction.LiveBench | 91.7% | — |

## Writing & Preference

- Gemini 2.5 Pro: 63.7 (#62)
- Gemini 3 Flash Preview: 65.5 (#45)

| Benchmark | Gemini 2.5 Pro | Gemini 3 Flash Preview |
|---|---|---|
| LMArena Text | 1458 | 1466 |
| LMArena Creative Writing | 1454 | 1457 |
| LMArena Multi-Turn | 1453 | 1471 |
| Short-Story Creative Writing | 83.8% | — |
| EQ-Bench Creative Writing | 1421 | — |
| WildBench | 85.7% | — |
| LiveBench Language | 67.8% | — |

## FAQ

### Is Gemini 2.5 Pro better than Gemini 3 Flash Preview?

Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 45.0 on the Noometry Index.

### Which is cheaper, Gemini 2.5 Pro or Gemini 3 Flash Preview?

Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; Gemini 2.5 Pro lists at $1.25 and $10.

### Is Gemini 2.5 Pro or Gemini 3 Flash Preview better for coding?

Gemini 3 Flash Preview scores higher on coding benchmarks: 50.9 versus 42.4 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

### How many benchmarks do Gemini 2.5 Pro and Gemini 3 Flash Preview share?

48 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and Gemini 3 Flash Preview has 59.