# GPT-4.1 mini vs GPT-6 Luna

> GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-1-mini-vs-gpt-6-luna
- Last updated: 2026-10-10
- Shared benchmarks: 31

## Summary

- They share 31 benchmarks with published results for both. GPT-4.1 mini scores higher in 0 categories and GPT-6 Luna in 10 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 24.1.
- The biggest single-benchmark swing is ARC-AGI-1: 3.5% for GPT-4.1 mini and 86.7% for GPT-6 Luna.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $0.40 / $1.60 for GPT-4.1 mini.
- GPT-6 Luna accepts more context: 1.05M tokens versus 1.05M.

## Snapshot

| | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 33.6 | 53.3 |
| Rank | 240 | 36 |
| Context | 1.05M | 1.05M |
| Input $/M | $0.40 | $0.10 |
| Output $/M | $1.60 | $0.50 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4.1 mini: 30.6 (#293)
- GPT-6 Luna: 55.5 (#25)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| SciCode | 40.4% | 54.6% |
| LMArena Coding | 1367 | 1439 |
| DeepSWE | — | 66.6% |
| FrontierCode | — | 42.4% |
| SWE-bench Verified (bash only) | 23.9% | — |
| Aider Polyglot | 32.4% | — |
| LMArena WebDev | — | 1581 |
| WeirdML | 37.6% | — |
| BigCodeBench Instruct | 48.9% | — |
| CadEval | 16% | — |
| ALE-Bench | — | 1,577 |

## Agentic & Tool Use

- GPT-4.1 mini: 33.3 (#55)
- GPT-6 Luna: 33.3 (#54)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| APEX-Agents | — | 44.3% |
| Berkeley Function Calling Leaderboard | 50.5% | — |
| GDP.pdf | — | 23% |

## Reasoning

- GPT-4.1 mini: 10.8 (#340)
- GPT-6 Luna: 48.2 (#41)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| ARC-AGI-2 | 0% | 59.3% |
| ARC-AGI-1 | 3.5% | 86.7% |
| CritPt | 0% | 19.4% |
| Chess Puzzles | 7% | 31% |
| LMArena Hard Prompts | 1349 | 1411 |
| Mystery Game Puzzles | 7% | 7% |
| DTBench | 68.8% | 90.1% |
| LMCA | 21.1% | 44.5% |
| Epoch Capabilities Index | 135.01 | 156.28 |
| Kagi LLM Benchmark | 48.6% | — |
| NYT Connections (extended) | — | 68.7% |

## Math

- GPT-4.1 mini: 24.1 (#270)
- GPT-6 Luna: 76.1 (#15)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6.7% | 78.9% |
| OTIS Mock AIME 2024-2025 | 44.7% | 98.9% |
| LMArena Math | 1343 | 1416 |
| FrontierMath Tier 4 | — | 56.1% |
| ProofBench | — | 64% |
| Omni-MATH | 49.1% | — |
| MATH Level 5 | 87.3% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |

## Knowledge

- GPT-4.1 mini: 34.7 (#194)
- GPT-6 Luna: 57.0 (#41)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| GPQA Diamond | 65.8% | 90.5% |
| SimpleQA Verified | 12.7% | 41.4% |
| LMArena Expert | 1338 | 1444 |
| MMLU-Pro | 78.3% | — |
| GPQA (HELM) | 61.4% | — |

## Multimodal

- GPT-4.1 mini: 35.8 (#82)
- GPT-6 Luna: 42.4 (#30)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| LMArena Vision | 1181 | 1217 |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 44.2% |

## Multilingual

- GPT-4.1 mini: 45.7 (#166)
- GPT-6 Luna: 50.5 (#117)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 1318 | 1386 |
| LMArena Chinese | 1329 | 1433 |
| LMArena French | 1358 | 1420 |
| LMArena German | 1351 | 1369 |
| LMArena Japanese | 1290 | 1369 |
| LMArena Korean | 1298 | 1360 |
| LMArena Russian | 1324 | 1394 |
| LMArena Spanish | 1319 | 1393 |

## Instruction Following

- GPT-4.1 mini: 73.7 (#118)
- GPT-6 Luna: 74.3 (#99)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1333 | 1409 |
| IFEval | 90.4% | — |

## Long Context

- GPT-4.1 mini: 31.8 (#275)
- GPT-6 Luna: 43.0 (#111)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | 1344 | 1409 |
| Fiction.LiveBench | 44.4% | — |

## Writing & Preference

- GPT-4.1 mini: 48.6 (#199)
- GPT-6 Luna: 58.3 (#119)

| Benchmark | GPT-4.1 mini | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1340 | 1391 |
| LMArena Creative Writing | 1300 | 1363 |
| LMArena Multi-Turn | 1354 | 1396 |
| EQ-Bench Creative Writing | 1147 | — |
| WildBench | 83.8% | — |

## FAQ

### Is GPT-4.1 mini better than GPT-6 Luna?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 33.6 on the Noometry Index.

### Which is cheaper, GPT-4.1 mini or GPT-6 Luna?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; GPT-4.1 mini lists at $0.40 and $1.60.

### Is GPT-4.1 mini or GPT-6 Luna better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 30.6 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 1.05M.

### How many benchmarks do GPT-4.1 mini and GPT-6 Luna share?

31 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and GPT-6 Luna has 42.