# Gemini 3 Pro vs o4-mini

> Gemini 3 Pro is the stronger model overall, scoring 54.7 to 41.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-pro-vs-o4-mini
- Last updated: 2026-10-10
- Shared benchmarks: 48

## Summary

- They share 48 benchmarks with published results for both. Gemini 3 Pro scores higher in 9 categories and o4-mini in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3 Pro leads 52.5 to 24.6.
- The biggest single-benchmark swing is SimpleBench: 76.4% for Gemini 3 Pro and 38.7% for o4-mini.

## Snapshot

| | Gemini 3 Pro | o4-mini |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 54.7 | 41.6 |
| Rank | 23 | 119 |
| Context | — | 200K |
| Input $/M | — | $1.10 |
| Output $/M | — | $4.40 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3 Pro: 52.1 (#33)
- o4-mini: 41.6 (#110)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| SWE-bench Verified (bash only) | 74.2% | 45% |
| GSO | 18.6% | 3.6% |
| WeirdML | 69.9% | 52.6% |
| LMArena Coding | 1482 | 1369 |
| ALE-Bench | 1,177 | 826.17 |
| AlgoTune | 1.83 | 1.72 |
| SWE-bench Verified | 72.9% | — |
| Aider Polyglot | — | 72% |
| LMArena WebDev | 1439 | — |
| SWE-bench Multilingual | 68.7% | — |
| CadEval | — | 62% |

## Agentic & Tool Use

- Gemini 3 Pro: 40.5 (#21)
- o4-mini: 32.3 (#57)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 72.5% | 53.2% |
| GDPval | 40.3% | 25.3% |
| METR Time Horizons | 71% | 63.9% |
| Terminal-Bench | 69.4% | — |
| Remote Labor Index | 1.3% | — |
| τ²-bench Airline | 80.5% | — |
| τ²-bench Banking | 18% | — |
| τ²-bench Retail | 75.9% | — |
| τ²-bench Telecom | 91% | — |
| DeepResearch Bench | 46.3% | — |
| BALROG | 58.1% | — |
| LMArena Search | 1207 | — |
| Vending-Bench 2 | 5,478 | — |

## Reasoning

- Gemini 3 Pro: 52.5 (#28)
- o4-mini: 24.6 (#143)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| ARC-AGI-2 | 31.1% | 6.1% |
| SimpleBench | 76.4% | 38.7% |
| Kagi LLM Benchmark | 80.1% | 67.6% |
| ARC-AGI-1 | 75% | 58.7% |
| CritPt | 6.9% | 0.6% |
| Chess Puzzles | 31% | 26% |
| EnigmaEval | 18.2% | 9.2% |
| LMArena Hard Prompts | 1481 | 1351 |
| Epoch Capabilities Index | 152.96 | 145.63 |
| ForecastBench | 61.2 | 61.8 |
| NYT Connections (extended) | 94.4% | — |
| Mystery Game Puzzles | — | 5% |
| DTBench | — | 77.6% |
| LMCA | — | 26.5% |

## Math

- Gemini 3 Pro: 48.8 (#50)
- o4-mini: 40.2 (#79)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.4% | 81.7% |
| Omni-MATH | 55.5% | 72% |
| LMArena Math | 1475 | 1390 |
| FrontierMath (Feb 2025 set) | 37.6% | 24.8% |
| FrontierMath Tier 4 (v1) | 18.8% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| MathArena Final-Answer Competitions | 67% | — |
| ProofBench | 20% | — |
| MATH Level 5 | — | 97.8% |

## Knowledge

- Gemini 3 Pro: 64.3 (#15)
- o4-mini: 43.5 (#83)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| GPQA Diamond | 92.6% | 79.6% |
| Humanity's Last Exam | 37.5% | 18.1% |
| MMLU-Pro | 90.3% | 82% |
| Vectara Hallucination Rate | 13.6% | 18.6% |
| GPQA (HELM) | 80.3% | 73.5% |
| LMArena Expert | 1475 | 1344 |
| SimpleQA Verified | — | 19.6% |
| Confabulations | — | 15.8% |

## Multimodal

- Gemini 3 Pro: 56.9 (#1)
- o4-mini: 39.5 (#46)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| LMArena Vision | 1305 | 1194 |
| GeoBench | 84% | 64% |
| VPCT | 91% | 57.5% |
| LMArena Document | 1434 | — |

## Multilingual

- Gemini 3 Pro: 57.1 (#14)
- o4-mini: 47.2 (#149)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| LMArena Non-English | 1474 | 1336 |
| LMArena Chinese | 1522 | 1355 |
| LMArena French | 1492 | 1362 |
| LMArena German | 1517 | 1336 |
| LMArena Japanese | 1513 | 1308 |
| LMArena Korean | 1449 | 1313 |
| LMArena Russian | 1493 | 1334 |
| LMArena Spanish | 1468 | 1346 |

## Instruction Following

- Gemini 3 Pro: 76.4 (#42)
- o4-mini: 75.4 (#63)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| IFEval | 87.7% | 92.8% |
| LMArena Instruction Following | 1458 | 1321 |

## Long Context

- Gemini 3 Pro: 44.0 (#75)
- o4-mini: 45.5 (#31)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| LMArena Longer Query | 1472 | 1315 |
| Fiction.LiveBench | — | 77.8% |
| CL-bench | 15.8% | — |

## Writing & Preference

- Gemini 3 Pro: 66.6 (#34)
- o4-mini: 54.2 (#148)

| Benchmark | Gemini 3 Pro | o4-mini |
|---|---|---|
| LMArena Text | 1479 | 1353 |
| LMArena Creative Writing | 1482 | 1294 |
| WildBench | 85.9% | 85.4% |
| LMArena Multi-Turn | 1484 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 1525 | — |

## FAQ

### Is Gemini 3 Pro better than o4-mini?

Gemini 3 Pro is the stronger model overall, scoring 54.7 to 41.6 on the Noometry Index.

### Is Gemini 3 Pro or o4-mini better for coding?

Gemini 3 Pro scores higher on coding benchmarks: 52.1 versus 41.6 in the Noometry coding category.

### How many benchmarks do Gemini 3 Pro and o4-mini share?

48 benchmarks have published results for both models. Gemini 3 Pro has 67 scored results on Noometry and o4-mini has 60.