# Gemini 3.1 Flash Lite vs GPT-4.5

> Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.4 to 36.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-1-flash-lite-vs-gpt-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 8 categories and GPT-4.5 in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Gemini 3.1 Flash Lite leads 41.7 to 32.4.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 80% for Gemini 3.1 Flash Lite and 37.8% for GPT-4.5.

## Snapshot

| | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 40.4 | 36.9 |
| Rank | 141 | 200 |
| Context | 1.05M | — |
| Input $/M | $0.25 | — |
| Output $/M | $1.50 | — |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.1 Flash Lite: 37.5 (#179)
- GPT-4.5: 41.8 (#102)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| WeirdML | 52.2% | 39.4% |
| LMArena Coding | 1400 | 1397 |
| Aider Polyglot | — | 44.9% |
| LMArena WebDev | 1256 | — |
| SciCode | 41.9% | — |
| LiveBench Coding | — | 75.2% |
| ALE-Bench | 797.73 | — |

## Agentic & Tool Use

- Gemini 3.1 Flash Lite: 29.9 (#72)
- GPT-4.5: 27.6 (#92)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| Cybench | — | 17.5% |
| DeepResearch Bench | 37.3% | — |

## Reasoning

- Gemini 3.1 Flash Lite: 22.3 (#181)
- GPT-4.5: 14.1 (#320)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| EnigmaEval | 3% | 3.2% |
| LMArena Hard Prompts | 1407 | 1404 |
| Epoch Capabilities Index | 144.45 | 136.74 |
| ForecastBench | 54.4 | 61.7 |
| ARC-AGI-2 | — | 0.8% |
| SimpleBench | — | 34.5% |
| Kagi LLM Benchmark | 67.2% | — |
| NYT Connections (extended) | 8.2% | — |
| ARC-AGI-1 | — | 10.3% |
| CritPt | 1.1% | — |
| Chess Puzzles | 25% | — |
| Thematic Generalization | 63.3% | — |
| LiveBench Reasoning | — | 71.1% |
| DTBench | 76.8% | — |
| LiveBench Data Analysis | — | 64.3% |
| LMCA | 35% | — |
| LiveBench | — | 69% |

## Math

- Gemini 3.1 Flash Lite: 39.5 (#82)
- GPT-4.5: 31.0 (#200)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 80% | 37.8% |
| LMArena Math | 1427 | 1412 |
| FrontierMath (Tiers 1-3) | 27.7% | — |
| LiveBench Math | — | 69.3% |
| MATH Level 5 | — | 78.6% |

## Knowledge

- Gemini 3.1 Flash Lite: 41.7 (#95)
- GPT-4.5: 32.4 (#205)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| GPQA Diamond | 81.8% | 68.7% |
| Humanity's Last Exam | 8.6% | 5.4% |
| LMArena Expert | 1397 | 1394 |
| Confabulations | — | 13.6% |
| Vectara Hallucination Rate | 8.2% | — |

## Multimodal

- Gemini 3.1 Flash Lite: 38.7 (#57)
- GPT-4.5: 37.0 (#68)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| LMArena Vision | 1240 | 1195 |
| VPCT | — | 45% |

## Multilingual

- Gemini 3.1 Flash Lite: 52.5 (#82)
- GPT-4.5: 52.7 (#78)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1410 | 1413 |
| LMArena Chinese | 1461 | 1421 |
| LMArena French | 1423 | 1417 |
| LMArena German | 1429 | 1457 |
| LMArena Japanese | 1411 | 1416 |
| LMArena Korean | 1393 | 1392 |
| LMArena Russian | 1420 | 1419 |
| LMArena Spanish | 1420 | — |

## Instruction Following

- Gemini 3.1 Flash Lite: 72.9 (#126)
- GPT-4.5: 72.8 (#129)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| LMArena Instruction Following | 1376 | 1404 |
| LiveBench Instruction Following | — | 72.3% |

## Long Context

- Gemini 3.1 Flash Lite: 42.5 (#117)
- GPT-4.5: 40.4 (#150)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| LMArena Longer Query | 1394 | 1406 |
| Fiction.LiveBench | — | 63.9% |

## Writing & Preference

- Gemini 3.1 Flash Lite: 61.0 (#90)
- GPT-4.5: 57.1 (#129)

| Benchmark | Gemini 3.1 Flash Lite | GPT-4.5 |
|---|---|---|
| LMArena Text | 1416 | 1417 |
| LMArena Creative Writing | 1401 | 1394 |
| LMArena Multi-Turn | 1417 | 1444 |
| Short-Story Creative Writing | — | 75.6% |
| EQ-Bench Creative Writing | — | 1258 |
| LiveBench Language | — | 61.5% |

## FAQ

### Is Gemini 3.1 Flash Lite better than GPT-4.5?

Gemini 3.1 Flash Lite is the stronger model overall, scoring 40.4 to 36.9 on the Noometry Index.

### Is Gemini 3.1 Flash Lite or GPT-4.5 better for coding?

GPT-4.5 scores higher on coding benchmarks: 41.8 versus 37.5 in the Noometry coding category.

### How many benchmarks do Gemini 3.1 Flash Lite and GPT-4.5 share?

24 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and GPT-4.5 has 42.