# Gemini 3.1 Pro Preview vs GPT-4.1

> Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 35.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-1-pro-preview-vs-gpt-4-1
- Last updated: 2026-10-10
- Shared benchmarks: 39

## Summary

- They share 39 benchmarks with published results for both. Gemini 3.1 Pro Preview scores higher in 10 categories and GPT-4.1 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemini 3.1 Pro Preview leads 71.7 to 11.7.
- The biggest single-benchmark swing is ARC-AGI-1: 98% for Gemini 3.1 Pro Preview and 5.5% for GPT-4.1.
- GPT-4.1 is cheaper at $2 / $8 per million input/output tokens, against $2 / $12 for Gemini 3.1 Pro Preview.
- Gemini 3.1 Pro Preview accepts more context: 1.05M tokens versus 1.05M.

## Snapshot

| | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 56.7 | 35.7 |
| Rank | 20 | 215 |
| Context | 1.05M | 1.05M |
| Input $/M | $2 | $2 |
| Output $/M | $12 | $8 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.1 Pro Preview: 42.7 (#88)
- GPT-4.1: 34.2 (#228)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| SWE-bench Verified | 75.6% | 48.5% |
| WeirdML | 72.1% | 39% |
| LMArena Coding | 1483 | 1391 |
| ALE-Bench | 1,161 | 558.1 |
| DeepSWE | 11.7% | — |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| LMArena WebDev | 1446 | — |
| SciCode | 58.9% | — |
| GSO | 22.6% | — |
| MirrorCode | 8.9% | — |
| CadEval | — | 42% |
| AlgoTune | 2.02 | — |

## Agentic & Tool Use

- Gemini 3.1 Pro Preview: 37.7 (#33)
- GPT-4.1: 34.4 (#40)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| Terminal-Bench | 80.2% | — |
| APEX-Agents | 35.3% | — |
| Berkeley Function Calling Leaderboard | — | 54% |
| τ²-bench Banking | 26% | — |
| DeepResearch Bench | 47.8% | — |
| PostTrainBench | 22% | — |
| BALROG | 57% | — |
| ExploitBench | 26.1% | — |
| GBAEval | 0.8% | — |
| GDP.pdf | 17% | — |
| LMArena Search | 1211 | — |
| METR Time Horizons | 77% | — |
| Vending-Bench 2 | 3,774 | — |

## Reasoning

- Gemini 3.1 Pro Preview: 71.7 (#9)
- GPT-4.1: 11.7 (#334)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| ARC-AGI-2 | 77.1% | 0.4% |
| SimpleBench | 79.6% | 27% |
| ARC-AGI-1 | 98% | 5.5% |
| Chess Puzzles | 55% | 6% |
| EnigmaEval | 36.8% | 2.2% |
| LMArena Hard Prompts | 1484 | 1384 |
| DTBench | 97.1% | 68.3% |
| LMCA | 53.8% | 25.6% |
| Epoch Capabilities Index | 154.92 | 136.79 |
| ForecastBench | 59 | 61.5 |
| Kagi LLM Benchmark | — | 52.3% |
| NYT Connections (extended) | 97.4% | — |
| CritPt | 17.7% | — |
| Thematic Generalization | 79.4% | — |
| EBR-Bench | 14.3% | — |
| Mystery Game Puzzles | 34% | — |

## Math

- Gemini 3.1 Pro Preview: 61.8 (#28)
- GPT-4.1: 21.4 (#269)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 59.6% | 6% |
| OTIS Mock AIME 2024-2025 | 95.6% | 38.3% |
| LMArena Math | 1485 | 1369 |
| FrontierMath (Feb 2025 set) | 36.9% | 5.5% |
| FrontierMath Tier 4 (v1) | 16.7% | 0% |
| FrontierMath Tier 4 | 26.8% | — |
| MathArena Final-Answer Competitions | 86.5% | — |
| ProofBench | 26% | — |
| Omni-MATH | — | 47.1% |
| MATH Level 5 | — | 83% |

## Knowledge

- Gemini 3.1 Pro Preview: 71.8 (#3)
- GPT-4.1: 37.0 (#152)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 94.4% | 66.9% |
| Humanity's Last Exam | 46.4% | 5.4% |
| SimpleQA Verified | 73.5% | 31.1% |
| Vectara Hallucination Rate | 10.4% | 5.6% |
| LMArena Expert | 1484 | 1364 |
| MMLU-Pro | — | 81.1% |
| GPQA (HELM) | — | 65.9% |

## Multimodal

- Gemini 3.1 Pro Preview: 38.1 (#60)
- GPT-4.1: 37.5 (#66)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| LMArena Vision | 1296 | 1210 |
| GeoBench | — | 72% |
| Blueprint-Bench 2 | 26.5% | — |
| Furniture Assembly | 26.7% | — |
| LMArena Document | 1444 | — |

## Multilingual

- Gemini 3.1 Pro Preview: 57.1 (#11)
- GPT-4.1: 49.6 (#128)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1475 | 1370 |
| LMArena Chinese | 1528 | 1382 |
| LMArena French | 1488 | 1383 |
| LMArena German | 1491 | 1380 |
| LMArena Japanese | 1494 | 1320 |
| LMArena Korean | 1458 | 1339 |
| LMArena Russian | 1496 | 1377 |
| LMArena Spanish | 1480 | 1375 |

## Instruction Following

- Gemini 3.1 Pro Preview: 77.1 (#31)
- GPT-4.1: 71.5 (#148)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| LMArena Instruction Following | 1465 | 1367 |
| IFEval | — | 83.8% |

## Long Context

- Gemini 3.1 Pro Preview: 47.3 (#17)
- GPT-4.1: 40.0 (#158)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1482 | 1385 |
| Fiction.LiveBench | — | 63.9% |
| CL-bench | 20.8% | — |
| CL-bench Life | 16.9% | — |

## Writing & Preference

- Gemini 3.1 Pro Preview: 66.3 (#36)
- GPT-4.1: 57.8 (#120)

| Benchmark | Gemini 3.1 Pro Preview | GPT-4.1 |
|---|---|---|
| LMArena Text | 1480 | 1383 |
| LMArena Creative Writing | 1482 | 1363 |
| EQ-Bench Creative Writing | 1491 | 1420 |
| LMArena Multi-Turn | 1486 | 1398 |
| WildBench | — | 85.4% |
| EQ-Bench 4 | 1142 | — |

## FAQ

### Is Gemini 3.1 Pro Preview better than GPT-4.1?

Gemini 3.1 Pro Preview is the stronger model overall, scoring 56.7 to 35.7 on the Noometry Index.

### Which is cheaper, Gemini 3.1 Pro Preview or GPT-4.1?

GPT-4.1 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; Gemini 3.1 Pro Preview lists at $2 and $12.

### Is Gemini 3.1 Pro Preview or GPT-4.1 better for coding?

Gemini 3.1 Pro Preview scores higher on coding benchmarks: 42.7 versus 34.2 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.1 Pro Preview does, with 1.05M tokens against 1.05M.

### How many benchmarks do Gemini 3.1 Pro Preview and GPT-4.1 share?

39 benchmarks have published results for both models. Gemini 3.1 Pro Preview has 71 scored results on Noometry and GPT-4.1 has 52.