# Gemini 2.5 Pro vs GPT-5.5

> GPT-5.5 is the stronger model overall, scoring 63.4 to 44.9 on the Noometry Index. Gemini 2.5 Pro costs 3.3× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-2-5-pro-vs-gpt-5-5
- Last updated: 2026-10-10
- Shared benchmarks: 49

## Summary

- They share 49 benchmarks with published results for both. Gemini 2.5 Pro scores higher in 1 category and GPT-5.5 in 9 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.5 leads 81.5 to 31.9.
- The biggest single-benchmark swing is ARC-AGI-2: 4.9% for Gemini 2.5 Pro and 85% for GPT-5.5.
- Gemini 2.5 Pro is cheaper at $1.25 / $10 per million input/output tokens, against $5 / $30 for GPT-5.5.
- GPT-5.5 accepts more context: 1.05M tokens versus 1.05M.

## Snapshot

| | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 44.9 | 63.4 |
| Rank | 67 | 8 |
| Context | 1.05M | 1.05M |
| Input $/M | $1.25 | $5 |
| Output $/M | $10 | $30 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 2.5 Pro: 42.8 (#86)
- GPT-5.5: 58.3 (#13)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| SWE-bench Verified | 57.6% | 80.6% |
| LMArena WebDev | 1227 | 1512 |
| SciCode | 42.8% | 56.1% |
| GSO | 3.9% | 40.2% |
| WeirdML | 54% | 84.9% |
| LMArena Coding | 1452 | 1494 |
| ALE-Bench | 785.52 | 1,943 |
| DeepSWE | — | 67% |
| FrontierCode | — | 43% |
| SWE-bench Verified (bash only) | 53.6% | — |
| Aider Polyglot | 83.1% | — |
| LiveBench Coding | 85.9% | — |
| MirrorCode | — | 10% |
| CadEval | 64% | — |
| AlgoTune | 1.51 | — |

## Agentic & Tool Use

- Gemini 2.5 Pro: 29.1 (#81)
- GPT-5.5: 50.7 (#6)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| Terminal-Bench | 32.6% | 84.7% |
| Remote Labor Index | 0.8% | 6.3% |
| τ²-bench Banking | 13.7% | 44.6% |
| DeepResearch Bench | 42.8% | 54% |
| LMArena Search | 1142 | 1242 |
| Vending-Bench 2 | 573.64 | 7,524 |
| APEX-Agents | — | 55.1% |
| OSWorld 2.0 | — | 13% |
| GDPval | 23.3% | — |
| TheAgentCompany | 30.3% | — |
| PostTrainBench | — | 27.2% |
| BALROG | 43.3% | — |
| ExploitBench | — | 47.4% |
| GBAEval | — | 53.2% |
| GDP.pdf | — | 26% |
| METR Time Horizons | 55.4% | — |

## Reasoning

- Gemini 2.5 Pro: 28.8 (#85)
- GPT-5.5: 72.7 (#8)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| ARC-AGI-2 | 4.9% | 85% |
| SimpleBench | 62.4% | 69% |
| Kagi LLM Benchmark | 70.3% | 88.8% |
| ARC-AGI-1 | 41% | 95% |
| CritPt | 2% | 27.1% |
| Chess Puzzles | 20% | 54% |
| LMArena Hard Prompts | 1455 | 1489 |
| DTBench | 82.4% | 96% |
| LMCA | 34.8% | 54.3% |
| Epoch Capabilities Index | 145.32 | 159.25 |
| ForecastBench | 61.3 | 60.6 |
| NYT Connections (extended) | — | 96.2% |
| EnigmaEval | 5.6% | — |
| EBR-Bench | — | 34.3% |
| LiveBench Reasoning | 89.8% | — |
| Mystery Game Puzzles | — | 56% |
| LiveBench Data Analysis | 79.9% | — |
| Surface Evolver Bench | — | 88.1% |
| Bench to the Future 3 | — | 0.14 |
| LiveBench | 82.3% | — |

## Math

- Gemini 2.5 Pro: 31.9 (#191)
- GPT-5.5: 81.5 (#9)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 24.6% | 85.3% |
| FrontierMath Tier 4 | 0% | 72.5% |
| OTIS Mock AIME 2024-2025 | 84.7% | 100% |
| LMArena Math | 1450 | 1486 |
| FrontierMath (Feb 2025 set) | 14.1% | 51.7% |
| FrontierMath Tier 4 (v1) | 4.2% | 35.4% |
| MathArena Final-Answer Competitions | — | 94.3% |
| ProofBench | — | 50% |
| Omni-MATH | 41.6% | — |
| LiveBench Math | 90.2% | — |
| MATH Level 5 | 95.9% | — |
| FrontierMath Erdős | — | 0% |

## Knowledge

- Gemini 2.5 Pro: 55.8 (#41)
- GPT-5.5: 64.4 (#14)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 85.3% | 94% |
| Vectara Hallucination Rate | 7% | 9.3% |
| LMArena Expert | 1452 | 1508 |
| Humanity's Last Exam | 21.6% | — |
| SimpleQA Verified | — | 63% |
| MMLU-Pro | 86.3% | — |
| Confabulations | 10.6% | — |
| GPQA (HELM) | 74.9% | — |

## Multimodal

- Gemini 2.5 Pro: 44.5 (#16)
- GPT-5.5: 47.1 (#11)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1263 | 1297 |
| LMArena Document | 1421 | 1486 |
| GeoBench | 86% | — |
| VPCT | 48% | — |
| Blueprint-Bench 2 | — | 36.2% |
| Furniture Assembly | — | 44.2% |
| SpatialViz-Bench | 44.7% | — |

## Multilingual

- Gemini 2.5 Pro: 55.5 (#30)
- GPT-5.5: 56.5 (#19)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1452 | 1467 |
| LMArena Chinese | 1507 | 1535 |
| LMArena French | 1473 | 1486 |
| LMArena German | 1487 | 1478 |
| LMArena Japanese | 1462 | 1498 |
| LMArena Korean | 1435 | 1461 |
| LMArena Russian | 1460 | 1473 |
| LMArena Spanish | 1473 | 1469 |

## Instruction Following

- Gemini 2.5 Pro: 75.2 (#71)
- GPT-5.5: 77.7 (#18)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1436 | 1479 |
| LiveBench Instruction Following | 80.6% | — |
| IFEval | 84% | — |

## Long Context

- Gemini 2.5 Pro: 59.7 (#5)
- GPT-5.5: 48.3 (#12)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1449 | 1484 |
| Fiction.LiveBench | 91.7% | — |
| CL-bench Life | — | 22.2% |

## Writing & Preference

- Gemini 2.5 Pro: 63.9 (#60)
- GPT-5.5: 72.8 (#13)

| Benchmark | Gemini 2.5 Pro | GPT-5.5 |
|---|---|---|
| LMArena Text | 1458 | 1471 |
| LMArena Creative Writing | 1453 | 1455 |
| EQ-Bench Creative Writing | 1421 | 1844 |
| LMArena Multi-Turn | 1453 | 1477 |
| Short-Story Creative Writing | 83.8% | — |
| WildBench | 85.7% | — |
| EQ-Bench 4 | — | 1315 |
| LiveBench Language | 67.8% | — |

## FAQ

### Is Gemini 2.5 Pro better than GPT-5.5?

GPT-5.5 is the stronger model overall, scoring 63.4 to 44.9 on the Noometry Index. Gemini 2.5 Pro costs 3.3× less per token, which makes it the better buy when GPT-5.5's lead doesn't matter for your workload.

### Which is cheaper, Gemini 2.5 Pro or GPT-5.5?

Gemini 2.5 Pro is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; GPT-5.5 lists at $5 and $30.

### Is Gemini 2.5 Pro or GPT-5.5 better for coding?

GPT-5.5 scores higher on coding benchmarks: 58.3 versus 42.8 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.5 does, with 1.05M tokens against 1.05M.

### How many benchmarks do Gemini 2.5 Pro and GPT-5.5 share?

49 benchmarks have published results for both models. Gemini 2.5 Pro has 78 scored results on Noometry and GPT-5.5 has 71.