# Gemini 3.1 Flash Lite vs GPT-5 Mini

> GPT-5 Mini is the stronger model overall, scoring 41.7 to 40.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemini-3-1-flash-lite-vs-gpt-5-mini
- Last updated: 2026-10-10
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. Gemini 3.1 Flash Lite scores higher in 4 categories and GPT-5 Mini in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 39.5.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 27.7% for Gemini 3.1 Flash Lite and 46.7% for GPT-5 Mini.
- Gemini 3.1 Flash Lite is cheaper at $0.25 / $1.50 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
- Gemini 3.1 Flash Lite accepts more context: 1.05M tokens versus 400K.

## Snapshot

| | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| Provider | Google | OpenAI |
| Noometry Index | 40.4 | 41.7 |
| Rank | 141 | 115 |
| Context | 1.05M | 400K |
| Input $/M | $0.25 | $0.25 |
| Output $/M | $1.50 | $2 |
| Weights | Proprietary | Proprietary |

## Coding

- Gemini 3.1 Flash Lite: 37.5 (#179)
- GPT-5 Mini: 40.0 (#134)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| SciCode | 41.9% | 39.2% |
| WeirdML | 52.2% | 52.7% |
| LMArena Coding | 1400 | 1406 |
| ALE-Bench | 797.73 | 799.77 |
| SWE-bench Verified | — | 64.7% |
| SWE-bench Verified (bash only) | — | 59.8% |
| LMArena WebDev | 1256 | — |
| SWE-bench Multilingual | — | 39.7% |
| AlgoTune | — | 1.38 |

## Agentic & Tool Use

- Gemini 3.1 Flash Lite: 29.9 (#72)
- GPT-5 Mini: 31.0 (#64)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | — | 34.8% |
| Berkeley Function Calling Leaderboard | — | 55.5% |
| DeepResearch Bench | 37.3% | — |
| Vending-Bench 2 | — | -31.18 |

## Reasoning

- Gemini 3.1 Flash Lite: 22.3 (#181)
- GPT-5 Mini: 24.1 (#153)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| Kagi LLM Benchmark | 67.2% | 70.3% |
| CritPt | 1.1% | 0% |
| Chess Puzzles | 25% | 30% |
| EnigmaEval | 3% | 8.2% |
| LMArena Hard Prompts | 1407 | 1380 |
| DTBench | 76.8% | 80.5% |
| LMCA | 35% | 34.2% |
| Epoch Capabilities Index | 144.45 | 145.51 |
| ForecastBench | 54.4 | 61 |
| ARC-AGI-2 | — | 4.4% |
| NYT Connections (extended) | 8.2% | — |
| ARC-AGI-1 | — | 54.3% |
| Thematic Generalization | 63.3% | — |
| Mystery Game Puzzles | — | 10% |

## Math

- Gemini 3.1 Flash Lite: 39.5 (#82)
- GPT-5 Mini: 46.3 (#59)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 27.7% | 46.7% |
| OTIS Mock AIME 2024-2025 | 80% | 86.7% |
| LMArena Math | 1427 | 1377 |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 9% |
| Omni-MATH | — | 72.2% |
| MATH Level 5 | — | 97.8% |
| FrontierMath (Feb 2025 set) | — | 27.2% |
| FrontierMath Tier 4 (v1) | — | 6.3% |

## Knowledge

- Gemini 3.1 Flash Lite: 41.7 (#95)
- GPT-5 Mini: 45.5 (#80)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 81.8% | 75% |
| Humanity's Last Exam | 8.6% | 19.4% |
| Vectara Hallucination Rate | 8.2% | 12.9% |
| LMArena Expert | 1397 | 1377 |
| SimpleQA Verified | — | 21.6% |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| GPQA (HELM) | — | 75.6% |

## Multimodal

- Gemini 3.1 Flash Lite: 38.7 (#57)
- GPT-5 Mini: 35.0 (#80)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| LMArena Vision | 1240 | 1202 |
| VPCT | — | 40.2% |

## Multilingual

- Gemini 3.1 Flash Lite: 52.5 (#82)
- GPT-5 Mini: 49.1 (#131)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1410 | 1363 |
| LMArena Chinese | 1461 | 1387 |
| LMArena French | 1423 | 1385 |
| LMArena German | 1429 | 1366 |
| LMArena Japanese | 1411 | 1342 |
| LMArena Korean | 1393 | 1306 |
| LMArena Russian | 1420 | 1362 |
| LMArena Spanish | 1420 | 1355 |

## Instruction Following

- Gemini 3.1 Flash Lite: 72.9 (#126)
- GPT-5 Mini: 76.4 (#43)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1376 | 1357 |
| IFEval | — | 92.7% |

## Long Context

- Gemini 3.1 Flash Lite: 42.5 (#117)
- GPT-5 Mini: 41.9 (#127)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1394 | 1355 |
| Fiction.LiveBench | — | 69.4% |

## Writing & Preference

- Gemini 3.1 Flash Lite: 61.0 (#90)
- GPT-5 Mini: 55.4 (#143)

| Benchmark | Gemini 3.1 Flash Lite | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1416 | 1373 |
| LMArena Creative Writing | 1401 | 1325 |
| LMArena Multi-Turn | 1417 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| EQ-Bench Creative Writing | — | 1313 |
| WildBench | — | 85.5% |

## FAQ

### Is Gemini 3.1 Flash Lite better than GPT-5 Mini?

GPT-5 Mini is the stronger model overall, scoring 41.7 to 40.4 on the Noometry Index.

### Which is cheaper, Gemini 3.1 Flash Lite or GPT-5 Mini?

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 Mini lists at $0.25 and $2.

### Is Gemini 3.1 Flash Lite or GPT-5 Mini better for coding?

GPT-5 Mini scores higher on coding benchmarks: 40.0 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 Mini share?

34 benchmarks have published results for both models. Gemini 3.1 Flash Lite has 38 scored results on Noometry and GPT-5 Mini has 60.