# Gemini 3 Flash Preview vs o3

> Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 47.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-flash-preview-vs-o3
- Last updated: 2026-10-10
- Shared benchmarks: 42

## Summary

- They share 42 benchmarks with published results for both. Gemini 3 Flash Preview scores higher in 9 categories and o3 in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Flash Preview leads 49.2 to 32.1.
- The biggest single-benchmark swing is ARC-AGI-2: 33.6% for Gemini 3 Flash Preview and 6.5% for o3.
- Gemini 3 Flash Preview is cheaper at $0.50 / $3 per million input/output tokens, against $2 / $8 for o3.
- Gemini 3 Flash Preview accepts more context: 1.05M tokens versus 200K.

## Snapshot

| | Gemini 3 Flash Preview | o3 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 52.3 | 47.4 |
| Rank | 33 | 53 |
| Context | 1.05M | 200K |
| Input $/M | $0.50 | $2 |
| Output $/M | $3 | $8 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3 Flash Preview: 51.5 (#35)
- o3: 47.3 (#57)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| SWE-bench Verified | 75.4% | 62.3% |
| SWE-bench Verified (bash only) | 75.8% | 58.4% |
| GSO | 9.8% | 8.8% |
| WeirdML | 61.6% | 52.4% |
| LMArena Coding | 1460 | 1408 |
| ALE-Bench | 1,367 | 933.55 |
| Aider Polyglot | — | 81.3% |
| LMArena WebDev | 1439 | — |
| SWE-bench Multilingual | 72.7% | — |
| CadEval | — | 74% |

## Agentic & Tool Use

- Gemini 3 Flash Preview: 38.6 (#27)
- o3: 34.3 (#42)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| DeepResearch Bench | 49.8% | 45.2% |
| LMArena Search | 1198 | 1144 |
| Terminal-Bench | 64.3% | — |
| Berkeley Function Calling Leaderboard | — | 63% |
| GDPval | — | 30.8% |
| τ²-bench Airline | 82.5% | — |
| τ²-bench Banking | 27.3% | — |
| τ²-bench Retail | 76.8% | — |
| τ²-bench Telecom | 91.2% | — |
| OSWorld | — | 23% |
| BALROG | 48.1% | — |
| GDP.pdf | 10% | — |
| METR Time Horizons | — | 65.4% |
| Vending-Bench 2 | 3,635 | — |

## Reasoning

- Gemini 3 Flash Preview: 49.2 (#35)
- o3: 32.1 (#66)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| ARC-AGI-2 | 33.6% | 6.5% |
| SimpleBench | 61.1% | 53.1% |
| ARC-AGI-1 | 84.7% | 60.8% |
| Chess Puzzles | 40% | 38% |
| LMArena Hard Prompts | 1465 | 1402 |
| Mystery Game Puzzles | 26% | 29% |
| DTBench | 89.1% | 84.8% |
| LMCA | 43.1% | 39.7% |
| Epoch Capabilities Index | 151.83 | 146.88 |
| ForecastBench | 58.5 | 62.5 |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 83.1% | — |
| CritPt | — | 1.4% |
| EnigmaEval | — | 13.1% |

## Math

- Gemini 3 Flash Preview: 51.4 (#42)
- o3: 48.9 (#49)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 51.2% | 33.3% |
| OTIS Mock AIME 2024-2025 | 95.6% | 84.4% |
| LMArena Math | 1472 | 1426 |
| FrontierMath (Feb 2025 set) | 35.6% | 18.7% |
| FrontierMath Tier 4 (v1) | 4.2% | 2.1% |
| FrontierMath Tier 4 | 17.1% | — |
| MathArena Final-Answer Competitions | 67.6% | — |
| ProofBench | 15% | — |
| Omni-MATH | — | 71.4% |
| MATH Level 5 | — | 97.8% |

## Knowledge

- Gemini 3 Flash Preview: 58.8 (#30)
- o3: 54.5 (#47)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| GPQA Diamond | 89.4% | 81.8% |
| SimpleQA Verified | 66.8% | 49.4% |
| LMArena Expert | 1461 | 1403 |
| Humanity's Last Exam | — | 20.3% |
| MMLU-Pro | — | 85.9% |
| Confabulations | — | 14.4% |
| Vectara Hallucination Rate | 13.5% | — |
| GPQA (HELM) | — | 75.3% |

## Multimodal

- Gemini 3 Flash Preview: 45.3 (#15)
- o3: 40.7 (#35)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| LMArena Vision | 1285 | 1214 |
| GeoBench | 88% | 74% |
| VPCT | 72.6% | 52% |
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1413 | — |

## Multilingual

- Gemini 3 Flash Preview: 55.9 (#26)
- o3: 51.9 (#99)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| LMArena Non-English | 1458 | 1402 |
| LMArena Chinese | 1510 | 1438 |
| LMArena French | 1478 | 1430 |
| LMArena German | 1497 | 1420 |
| LMArena Japanese | 1488 | 1404 |
| LMArena Korean | 1443 | 1371 |
| LMArena Russian | 1480 | 1407 |
| LMArena Spanish | 1470 | 1395 |

## Instruction Following

- Gemini 3 Flash Preview: 75.8 (#52)
- o3: 73.0 (#122)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| LMArena Instruction Following | 1437 | 1368 |
| IFEval | — | 86.9% |

## Long Context

- Gemini 3 Flash Preview: 44.4 (#63)
- o3: 53.2 (#6)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| LMArena Longer Query | 1452 | 1372 |
| Fiction.LiveBench | — | 88.9% |
| CL-bench | — | 17.8% |

## Writing & Preference

- Gemini 3 Flash Preview: 65.6 (#44)
- o3: 63.7 (#62)

| Benchmark | Gemini 3 Flash Preview | o3 |
|---|---|---|
| LMArena Text | 1466 | 1410 |
| LMArena Creative Writing | 1457 | 1359 |
| LMArena Multi-Turn | 1471 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
| EQ-Bench Creative Writing | — | 1676 |
| WildBench | — | 86.1% |

## FAQ

### Is Gemini 3 Flash Preview better than o3?

Gemini 3 Flash Preview is the stronger model overall, scoring 52.3 to 47.4 on the Noometry Index.

### Which is cheaper, Gemini 3 Flash Preview or o3?

Gemini 3 Flash Preview is cheaper. It lists at $0.50 per million input tokens and $3 per million output tokens; o3 lists at $2 and $8.

### Is Gemini 3 Flash Preview or o3 better for coding?

Gemini 3 Flash Preview scores higher on coding benchmarks: 51.5 versus 47.3 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3 Flash Preview does, with 1.05M tokens against 200K.

### How many benchmarks do Gemini 3 Flash Preview and o3 share?

42 benchmarks have published results for both models. Gemini 3 Flash Preview has 59 scored results on Noometry and o3 has 63.