# Gemini 3.1 Flash Lite vs GPT-5.2 Codex

> GPT-5.2 Codex is the stronger model overall, scoring 42.2 to 40.4 on the Noometry Index. Gemini 3.1 Flash Lite costs 8.6× less per token, which makes it the better buy when GPT-5.2 Codex's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gemini-3-1-flash-lite-vs-gpt-5-2-codex
- Last updated: 2026-10-10
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 0 categories and GPT-5.2 Codex in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.2 Codex leads 40.7 to 29.9.
- Gemini 3.1 Flash Lite is cheaper at $0.25 / $1.50 per million input/output tokens, against $1.75 / $14 for GPT-5.2 Codex.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 400K.

## Snapshot

| | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 40.4 | 42.2 |
| Rank | 141 | 107 |
| Context | 1.05M | 400K |
| Input $/M | $0.25 | $1.75 |
| Output $/M | $1.50 | $14 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.1 Flash Lite: 37.5 (#179)
- GPT-5.2 Codex: 45.4 (#66)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| LMArena WebDev | 1256 | 1339 |
| ALE-Bench | 797.73 | 1,300 |
| SWE-bench Verified (bash only) | — | 72.8% |
| SWE-bench Multilingual | — | 66.3% |
| SciCode | 41.9% | — |
| WeirdML | 52.2% | — |
| LMArena Coding | 1400 | — |

## Agentic & Tool Use

- Gemini 3.1 Flash Lite: 29.9 (#72)
- GPT-5.2 Codex: 40.7 (#20)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| Terminal-Bench | — | 66.5% |
| DeepResearch Bench | 37.3% | — |

## Reasoning

- Gemini 3.1 Flash Lite: 22.3 (#181)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | — |
| NYT Connections (extended) | 8.2% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | 25% | — |
| EnigmaEval | 3% | — |
| Thematic Generalization | 63.3% | — |
| LMArena Hard Prompts | 1407 | — |
| DTBench | 76.8% | — |
| LMCA | 35% | — |
| Epoch Capabilities Index | 144.45 | — |
| ForecastBench | 54.4 | — |

## Math

- Gemini 3.1 Flash Lite: 39.5 (#82)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| FrontierMath (Tiers 1-3) | 27.7% | — |
| OTIS Mock AIME 2024-2025 | 80% | — |
| LMArena Math | 1427 | — |

## Knowledge

- Gemini 3.1 Flash Lite: 41.7 (#95)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| GPQA Diamond | 81.8% | — |
| Humanity's Last Exam | 8.6% | — |
| Vectara Hallucination Rate | 8.2% | — |
| LMArena Expert | 1397 | — |

## Multimodal

- Gemini 3.1 Flash Lite: 38.7 (#57)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| LMArena Vision | 1240 | — |

## Multilingual

- Gemini 3.1 Flash Lite: 52.5 (#82)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| LMArena Non-English | 1410 | — |
| LMArena Chinese | 1461 | — |
| LMArena French | 1423 | — |
| LMArena German | 1429 | — |
| LMArena Japanese | 1411 | — |
| LMArena Korean | 1393 | — |
| LMArena Russian | 1420 | — |
| LMArena Spanish | 1420 | — |

## Instruction Following

- Gemini 3.1 Flash Lite: 72.9 (#126)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| LMArena Instruction Following | 1376 | — |

## Long Context

- Gemini 3.1 Flash Lite: 42.5 (#117)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| LMArena Longer Query | 1394 | — |

## Writing & Preference

- Gemini 3.1 Flash Lite: 61.0 (#90)
- GPT-5.2 Codex: —

| Benchmark | Gemini 3.1 Flash Lite | GPT-5.2 Codex |
|---|---|---|
| LMArena Text | 1416 | — |
| LMArena Creative Writing | 1401 | — |
| LMArena Multi-Turn | 1417 | — |

## FAQ

### Is Gemini 3.1 Flash Lite better than GPT-5.2 Codex?

GPT-5.2 Codex is the stronger model overall, scoring 42.2 to 40.4 on the Noometry Index. Gemini 3.1 Flash Lite costs 8.6× less per token, which makes it the better buy when GPT-5.2 Codex's lead doesn't matter for your workload.

### Which is cheaper, Gemini 3.1 Flash Lite or GPT-5.2 Codex?

Gemini 3.1 Flash Lite is cheaper. It lists at $0.25 per million input tokens and $1.50 per million output tokens; GPT-5.2 Codex lists at $1.75 and $14.

### Is Gemini 3.1 Flash Lite or GPT-5.2 Codex better for coding?

GPT-5.2 Codex scores higher on coding benchmarks: 45.4 versus 37.5 in the Noometry coding category.

### Which has the bigger context window?

Gemini 3.1 Flash Lite does, with 1.05M tokens against 400K.

### How many benchmarks do Gemini 3.1 Flash Lite and GPT-5.2 Codex share?

2 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and GPT-5.2 Codex has 5.