# GPT-4.1 mini vs GPT-4o

> GPT-4.1 mini is the stronger model overall, scoring 33.5 to 28.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-1-mini-vs-gpt-4o
- Last updated: 2026-10-10
- Shared benchmarks: 42

## Summary

- They share 42 benchmarks with published results for both. GPT-4.1 mini scores higher in 8 categories and GPT-4o in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-4.1 mini leads 23.1 to 10.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 44.7% for GPT-4.1 mini and 6.4% for GPT-4o.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 128K.

## Snapshot

| | GPT-4.1 mini | GPT-4o |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 33.5 | 28.5 |
| Rank | 232 | 316 |
| Context | 1.05M | 128K |
| Input $/M | $0.40 | $2.50 |
| Output $/M | $1.60 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4.1 mini: 30.4 (#286)
- GPT-4o: 25.1 (#323)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| SWE-bench Verified (bash only) | 23.9% | 21.6% |
| Aider Polyglot | 32.4% | 45.3% |
| WeirdML | 37.6% | 25.1% |
| BigCodeBench Instruct | 48.9% | 51.1% |
| LMArena Coding | 1367 | 1297 |
| CadEval | 16% | 26% |
| SWE-bench Verified | — | 31% |
| SciCode | 40.4% | — |
| GSO | — | 0% |
| LiveBench Coding | — | 51.4% |
| BigCodeBench Complete | — | 61.1% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |

## Agentic & Tool Use

- GPT-4.1 mini: 33.0 (#52)
- GPT-4o: 20.8 (#133)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| Berkeley Function Calling Leaderboard | 50.5% | — |
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| Cybench | — | 12.5% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |

## Reasoning

- GPT-4.1 mini: 10.9 (#335)
- GPT-4o: 9.4 (#338)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 0% | 0% |
| ARC-AGI-1 | 3.5% | 4.5% |
| CritPt | 0% | 0% |
| Chess Puzzles | 7% | 13% |
| LMArena Hard Prompts | 1349 | 1281 |
| DTBench | 68.8% | 64.5% |
| LMCA | 21.1% | 16.6% |
| Epoch Capabilities Index | 135.01 | 128.97 |
| SimpleBench | — | 17.8% |
| Kagi LLM Benchmark | 48.6% | — |
| EnigmaEval | — | 0.8% |
| LiveBench Reasoning | — | 55.8% |
| Mystery Game Puzzles | 7% | — |
| LiveBench Data Analysis | — | 60.9% |
| ForecastBench | — | 57.7 |
| LiveBench | — | 55.3% |

## Math

- GPT-4.1 mini: 23.1 (#264)
- GPT-4o: 10.2 (#303)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6.7% | 0.4% |
| OTIS Mock AIME 2024-2025 | 44.7% | 6.4% |
| Omni-MATH | 49.1% | 29.3% |
| LMArena Math | 1343 | 1285 |
| MATH Level 5 | 87.3% | 53.3% |
| FrontierMath (Feb 2025 set) | 4.5% | 0.3% |
| LiveBench Math | — | 49.5% |

## Knowledge

- GPT-4.1 mini: 34.7 (#183)
- GPT-4o: 28.8 (#235)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| GPQA Diamond | 65.8% | 49.2% |
| SimpleQA Verified | 12.7% | 26% |
| MMLU-Pro | 78.3% | 71.3% |
| GPQA (HELM) | 61.4% | 52% |
| LMArena Expert | 1338 | 1250 |
| Humanity's Last Exam | — | 2.7% |
| Confabulations | — | 15.3% |
| Vectara Hallucination Rate | — | 9.6% |
| MMLU | — | 88.1% |

## Multimodal

- GPT-4.1 mini: 35.2 (#77)
- GPT-4o: 33.9 (#85)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| LMArena Vision | 1181 | 1137 |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| VPCT | — | 40% |
| ScienceQA | — | 88.5% |

## Multilingual

- GPT-4.1 mini: 45.9 (#161)
- GPT-4o: 43.4 (#181)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| LMArena Non-English | 1318 | 1283 |
| LMArena Chinese | 1329 | 1277 |
| LMArena French | 1357 | 1303 |
| LMArena German | 1349 | 1282 |
| LMArena Japanese | 1291 | 1257 |
| LMArena Korean | 1298 | 1234 |
| LMArena Russian | 1324 | 1286 |
| LMArena Spanish | 1319 | 1292 |

## Instruction Following

- GPT-4.1 mini: 73.9 (#113)
- GPT-4o: 66.8 (#202)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| IFEval | 90.4% | 81.7% |
| LMArena Instruction Following | 1332 | 1278 |
| LiveBench Instruction Following | — | 68.6% |

## Long Context

- GPT-4.1 mini: 31.8 (#270)
- GPT-4o: 39.4 (#174)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| Fiction.LiveBench | 44.4% | 66.7% |
| LMArena Longer Query | 1344 | 1289 |

## Writing & Preference

- GPT-4.1 mini: 48.8 (#193)
- GPT-4o: 52.8 (#160)

| Benchmark | GPT-4.1 mini | GPT-4o |
|---|---|---|
| LMArena Text | 1341 | 1300 |
| LMArena Creative Writing | 1300 | 1291 |
| WildBench | 83.8% | 82.8% |
| LMArena Multi-Turn | 1354 | 1302 |
| Short-Story Creative Writing | — | 81.8% |
| EQ-Bench Creative Writing | 1147 | — |
| LiveBench Language | — | 47.6% |

## FAQ

### Is GPT-4.1 mini better than GPT-4o?

GPT-4.1 mini is the stronger model overall, scoring 33.5 to 28.5 on the Noometry Index.

### Which is cheaper, GPT-4.1 mini or GPT-4o?

GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; GPT-4o lists at $2.50 and $10.

### Is GPT-4.1 mini or GPT-4o better for coding?

GPT-4.1 mini scores higher on coding benchmarks: 30.4 versus 25.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-4.1 mini does, with 1.05M tokens against 128K.

### How many benchmarks do GPT-4.1 mini and GPT-4o share?

42 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and GPT-4o has 72.