# Gemini 3.5 Flash vs GPT-5.2

> Gemini 3.5 Flash and GPT-5.2 score almost the same on the Noometry Index (54.1 vs 54.1), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/gemini-3-5-flash-vs-gpt-5-2
- Last updated: 2026-10-10
- Shared benchmarks: 45

## Summary

- They share 45 benchmarks with published results for both. Gemini 3.5 Flash scores higher in 6 categories and GPT-5.2 in 4 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.2 leads 40.0 to 24.8.
- The biggest single-benchmark swing is SimpleBench: 76.7% for Gemini 3.5 Flash and 45.8% for GPT-5.2.
- Gemini 3.5 Flash is cheaper at $1.50 / $9 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
- Gemini 3.5 Flash accepts more context: 1.05M tokens versus 400K.

## Snapshot

| | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 54.1 | 54.1 |
| Rank | 29 | 31 |
| Context | 1.05M | 400K |
| Input $/M | $1.50 | $1.75 |
| Output $/M | $9 | $14 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.5 Flash: 49.1 (#46)
- GPT-5.2: 52.2 (#32)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| SWE-bench Verified | 79.3% | 73.8% |
| LMArena WebDev | 1499 | 1415 |
| WeirdML | 62.6% | 72.2% |
| LMArena Coding | 1492 | 1447 |
| ALE-Bench | 911.02 | 1,294 |
| DeepSWE | 37.4% | — |
| SWE-bench Verified (bash only) | — | 72.8% |
| SWE-bench Multilingual | — | 66.7% |
| SciCode | 53.1% | — |
| GSO | — | 27.4% |
| AlgoTune | — | 2.05 |

## Agentic & Tool Use

- Gemini 3.5 Flash: 24.8 (#107)
- GPT-5.2: 40.0 (#24)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| Vending-Bench 2 | 5,396 | 3,591 |
| Terminal-Bench | — | 64.9% |
| APEX-Agents | 27.5% | — |
| Berkeley Function Calling Leaderboard | — | 55.9% |
| GDPval | — | 49.7% |
| Remote Labor Index | — | 2.5% |
| τ²-bench Airline | — | 83% |
| τ²-bench Banking | — | 32.2% |
| τ²-bench Retail | — | 81.6% |
| τ²-bench Telecom | — | 89.7% |
| DeepResearch Bench | — | 41.1% |
| GBAEval | 6.7% | — |
| GDP.pdf | 14% | — |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 75.3% |

## Reasoning

- Gemini 3.5 Flash: 62.7 (#16)
- GPT-5.2: 50.1 (#33)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| ARC-AGI-2 | 72.1% | 52.9% |
| SimpleBench | 76.7% | 45.8% |
| NYT Connections (extended) | 92.6% | 83.6% |
| ARC-AGI-1 | 92.5% | 86.2% |
| Chess Puzzles | 50% | 49% |
| EnigmaEval | 25.4% | 10.4% |
| EBR-Bench | 4.8% | 23% |
| LMArena Hard Prompts | 1488 | 1445 |
| Mystery Game Puzzles | 32% | 23% |
| DTBench | 94.7% | 90.9% |
| LMCA | 47.1% | 43.9% |
| Epoch Capabilities Index | 154.55 | 153.45 |
| ForecastBench | 59 | 60.1 |
| Kagi LLM Benchmark | — | 73.3% |
| CritPt | 13.1% | — |
| Surface Evolver Bench | 58.1% | — |

## Math

- Gemini 3.5 Flash: 60.5 (#30)
- GPT-5.2: 59.8 (#32)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 62.8% | 67.4% |
| FrontierMath Tier 4 | 26.8% | 31.7% |
| MathArena Final-Answer Competitions | 76.3% | 72% |
| OTIS Mock AIME 2024-2025 | 95.6% | 96.1% |
| ProofBench | 31% | 15% |
| LMArena Math | 1502 | 1439 |
| FrontierMath (Feb 2025 set) | 39% | 40.7% |
| FrontierMath Tier 4 (v1) | 14.6% | 18.8% |

## Knowledge

- Gemini 3.5 Flash: 66.3 (#10)
- GPT-5.2: 59.2 (#29)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 92.8% | 91.4% |
| SimpleQA Verified | 66.2% | 37.1% |
| LMArena Expert | 1493 | 1445 |
| Humanity's Last Exam | — | 27.8% |
| Vectara Hallucination Rate | — | 8.4% |

## Multimodal

- Gemini 3.5 Flash: 45.7 (#13)
- GPT-5.2: 50.9 (#7)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1311 | 1268 |
| LMArena Document | 1463 | 1405 |
| VPCT | — | 84% |
| Blueprint-Bench 2 | 33.6% | — |
| Furniture Assembly | — | 38.3% |

## Multilingual

- Gemini 3.5 Flash: 57.1 (#12)
- GPT-5.2: 53.6 (#63)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1475 | 1425 |
| LMArena Chinese | 1525 | 1460 |
| LMArena French | 1490 | 1456 |
| LMArena German | 1500 | 1448 |
| LMArena Japanese | 1481 | 1421 |
| LMArena Korean | 1453 | 1390 |
| LMArena Russian | 1491 | 1440 |
| LMArena Spanish | 1478 | 1433 |

## Instruction Following

- Gemini 3.5 Flash: 77.2 (#29)
- GPT-5.2: 74.9 (#83)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1466 | 1417 |

## Long Context

- Gemini 3.5 Flash: 45.3 (#39)
- GPT-5.2: 44.0 (#74)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| LMArena Longer Query | 1481 | 1428 |
| CL-bench | — | 18.2% |

## Writing & Preference

- Gemini 3.5 Flash: 65.5 (#46)
- GPT-5.2: 67.0 (#30)

| Benchmark | Gemini 3.5 Flash | GPT-5.2 |
|---|---|---|
| LMArena Text | 1481 | 1439 |
| LMArena Creative Writing | 1467 | 1402 |
| LMArena Multi-Turn | 1482 | 1458 |
| EQ-Bench Creative Writing | — | 1703 |
| EQ-Bench 4 | 1087 | — |

## FAQ

### Is Gemini 3.5 Flash better than GPT-5.2?

Gemini 3.5 Flash and GPT-5.2 score almost the same on the Noometry Index (54.1 vs 54.1), so choose on price, context window or the category you care about most.

### Which is cheaper, Gemini 3.5 Flash or GPT-5.2?

Gemini 3.5 Flash is cheaper. It lists at $1.50 per million input tokens and $9 per million output tokens; GPT-5.2 lists at $1.75 and $14.

### Is Gemini 3.5 Flash or GPT-5.2 better for coding?

GPT-5.2 scores higher on coding benchmarks: 52.2 versus 49.1 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.5 Flash does, with 1.05M tokens against 400K.

### How many benchmarks do Gemini 3.5 Flash and GPT-5.2 share?

45 benchmarks have published results for both models. Gemini 3.5 Flash has 54 scored results on Noometry and GPT-5.2 has 67.