# GPT-4.1 vs GPT-4.1 mini

> GPT-4.1 is the stronger model overall, scoring 35.7 to 33.5 on the Noometry Index. GPT-4.1 mini costs 5.0× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-4-1-vs-gpt-4-1-mini
- Last updated: 2026-10-10
- Shared benchmarks: 43

## Summary

- They share 43 benchmarks with published results for both. GPT-4.1 scores higher in 8 categories and GPT-4.1 mini in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.1 leads 57.8 to 48.8.
- The biggest single-benchmark swing is CadEval: 42% for GPT-4.1 and 16% for GPT-4.1 mini.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $2 / $8 for GPT-4.1.

## Snapshot

| | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.7 | 33.5 |
| Rank | 215 | 232 |
| Context | 1.05M | 1.05M |
| Input $/M | $2 | $0.40 |
| Output $/M | $8 | $1.60 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4.1: 34.2 (#228)
- GPT-4.1 mini: 30.4 (#286)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| SWE-bench Verified (bash only) | 39.6% | 23.9% |
| Aider Polyglot | 52.4% | 32.4% |
| WeirdML | 39% | 37.6% |
| LMArena Coding | 1391 | 1367 |
| CadEval | 42% | 16% |
| SWE-bench Verified | 48.5% | — |
| SciCode | — | 40.4% |
| BigCodeBench Instruct | — | 48.9% |
| ALE-Bench | 558.1 | — |

## Agentic & Tool Use

- GPT-4.1: 34.4 (#40)
- GPT-4.1 mini: 33.0 (#52)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | 50.5% |

## Reasoning

- GPT-4.1: 11.7 (#334)
- GPT-4.1 mini: 10.9 (#335)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| ARC-AGI-2 | 0.4% | 0% |
| Kagi LLM Benchmark | 52.3% | 48.6% |
| ARC-AGI-1 | 5.5% | 3.5% |
| Chess Puzzles | 6% | 7% |
| LMArena Hard Prompts | 1384 | 1349 |
| DTBench | 68.3% | 68.8% |
| LMCA | 25.6% | 21.1% |
| Epoch Capabilities Index | 136.79 | 135.01 |
| SimpleBench | 27% | — |
| CritPt | — | 0% |
| EnigmaEval | 2.2% | — |
| Mystery Game Puzzles | — | 7% |
| ForecastBench | 61.5 | — |

## Math

- GPT-4.1: 21.4 (#269)
- GPT-4.1 mini: 23.1 (#264)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 6.7% |
| OTIS Mock AIME 2024-2025 | 38.3% | 44.7% |
| Omni-MATH | 47.1% | 49.1% |
| LMArena Math | 1369 | 1343 |
| MATH Level 5 | 83% | 87.3% |
| FrontierMath (Feb 2025 set) | 5.5% | 4.5% |
| FrontierMath Tier 4 (v1) | 0% | — |

## Knowledge

- GPT-4.1: 37.0 (#152)
- GPT-4.1 mini: 34.7 (#183)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 66.9% | 65.8% |
| SimpleQA Verified | 31.1% | 12.7% |
| MMLU-Pro | 81.1% | 78.3% |
| GPQA (HELM) | 65.9% | 61.4% |
| LMArena Expert | 1364 | 1338 |
| Humanity's Last Exam | 5.4% | — |
| Vectara Hallucination Rate | 5.6% | — |

## Multimodal

- GPT-4.1: 37.5 (#66)
- GPT-4.1 mini: 35.2 (#77)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | 1210 | 1181 |
| GeoBench | 72% | — |

## Multilingual

- GPT-4.1: 49.6 (#128)
- GPT-4.1 mini: 45.9 (#161)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1370 | 1318 |
| LMArena Chinese | 1382 | 1329 |
| LMArena French | 1383 | 1357 |
| LMArena German | 1380 | 1349 |
| LMArena Japanese | 1320 | 1291 |
| LMArena Korean | 1339 | 1298 |
| LMArena Russian | 1377 | 1324 |
| LMArena Spanish | 1375 | 1319 |

## Instruction Following

- GPT-4.1: 71.5 (#148)
- GPT-4.1 mini: 73.9 (#113)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| IFEval | 83.8% | 90.4% |
| LMArena Instruction Following | 1367 | 1332 |

## Long Context

- GPT-4.1: 40.0 (#158)
- GPT-4.1 mini: 31.8 (#270)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| Fiction.LiveBench | 63.9% | 44.4% |
| LMArena Longer Query | 1385 | 1344 |

## Writing & Preference

- GPT-4.1: 57.8 (#120)
- GPT-4.1 mini: 48.8 (#193)

| Benchmark | GPT-4.1 | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1383 | 1341 |
| LMArena Creative Writing | 1363 | 1300 |
| EQ-Bench Creative Writing | 1420 | 1147 |
| WildBench | 85.4% | 83.8% |
| LMArena Multi-Turn | 1398 | 1354 |

## FAQ

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

GPT-4.1 is the stronger model overall, scoring 35.7 to 33.5 on the Noometry Index. GPT-4.1 mini costs 5.0× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.

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

GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; GPT-4.1 lists at $2 and $8.

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

GPT-4.1 scores higher on coding benchmarks: 34.2 versus 30.4 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1.05M tokens.

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

43 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and GPT-4.1 mini has 47.