Model comparison
GPT-4.1 vs GPT-5.6 Luna
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 35.9 on the Noometry Index.
Last verified . 33 shared benchmarks.
Summary
- They share 33 benchmarks with published results for both. GPT-4.1 scores higher in 1 category and GPT-5.6 Luna in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 22.3.
- The biggest single-benchmark swing is ARC-AGI-1: 5.5% for GPT-4.1 and 88% for GPT-5.6 Luna.
- GPT-5.6 Luna is cheaper at $0.20 / $1.20 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-5.6 Luna accepts more context: 1.05M tokens versus 1.05M.
Side by side
| GPT-4.1 | GPT-5.6 Luna | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.9 | 54.6 |
| Released | 2025-04-14 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 33K | 128K |
| Input $ / M tokens | $2 | $0.20 |
| Output $ / M tokens | $8 | $1.20 |
| Results tracked | 52 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
GPT-4.1: 34.4 (#238), GPT-5.6 Luna: 54.5 (#28)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| WeirdML | 39% | 60.9% |
| LMArena Coding | 1391 | 1466 |
| ALE-Bench | 558.1 | 1,667 |
| SWE-bench Verified | 48.5% | — |
| DeepSWE | — | 67.2% |
| FrontierCode | — | 39.8% |
| SWE-bench Verified (bash only) | 39.6% | — |
| Aider Polyglot | 52.4% | — |
| CursorBench | — | 35.9% |
| LMArena WebDev | — | 1519 |
| SciCode | — | 53.6% |
| CadEval | 42% | — |
Agentic & Tool Use Too close to call
GPT-4.1: 34.7 (#43), GPT-5.6 Luna: 34.4 (#45)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| APEX-Agents | — | 43% |
| Berkeley Function Calling Leaderboard | 54% | — |
| BALROG | — | 45.6% |
| GDP.pdf | — | 22.7% |
| Vending-Bench 2 | — | 4,095 |
Reasoning GPT-5.6 Luna leads
GPT-4.1: 11.7 (#339), GPT-5.6 Luna: 47.6 (#43)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| ARC-AGI-2 | 0.4% | 59.5% |
| SimpleBench | 27% | 46.8% |
| Kagi LLM Benchmark | 52.3% | 49.1% |
| ARC-AGI-1 | 5.5% | 88% |
| Chess Puzzles | 6% | 40% |
| LMArena Hard Prompts | 1384 | 1451 |
| DTBench | 68.3% | 89.1% |
| LMCA | 25.6% | 48.5% |
| Epoch Capabilities Index | 136.78 | 156.39 |
| NYT Connections (extended) | — | 69.4% |
| CritPt | — | 20.6% |
| EnigmaEval | 2.2% | — |
| Mystery Game Puzzles | — | 21% |
| Surface Evolver Bench | — | 61.9% |
| ForecastBench | 61.5 | — |
Math GPT-5.6 Luna leads
GPT-4.1: 22.3 (#280), GPT-5.6 Luna: 77.7 (#14)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 82.1% |
| OTIS Mock AIME 2024-2025 | 38.3% | 98.3% |
| LMArena Math | 1370 | 1458 |
| FrontierMath Tier 4 | — | 61% |
| ProofBench | — | 60% |
| Omni-MATH | 47.1% | — |
| MATH Level 5 | 83% | — |
| FrontierMath (Feb 2025 set) | 5.5% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-5.6 Luna leads
GPT-4.1: 37.1 (#160), GPT-5.6 Luna: 58.5 (#34)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| GPQA Diamond | 66.9% | 91.6% |
| SimpleQA Verified | 31.1% | 41% |
| LMArena Expert | 1364 | 1478 |
| Humanity's Last Exam | 5.4% | — |
| MMLU-Pro | 81.1% | — |
| Vectara Hallucination Rate | 5.6% | — |
| GPQA (HELM) | 65.9% | — |
Multimodal GPT-5.6 Luna leads
GPT-4.1: 38.2 (#67), GPT-5.6 Luna: 42.7 (#28)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| LMArena Vision | 1211 | 1258 |
| GeoBench | 72% | — |
| Blueprint-Bench 2 | — | 22.6% |
| Furniture Assembly | — | 42.5% |
| LMArena Document | — | 1457 |
Multilingual GPT-5.6 Luna leads
GPT-4.1: 49.4 (#133), GPT-5.6 Luna: 52.8 (#78)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| LMArena Non-English | 1370 | 1417 |
| LMArena Chinese | 1382 | 1470 |
| LMArena French | 1382 | 1456 |
| LMArena German | 1381 | 1454 |
| LMArena Japanese | 1319 | 1411 |
| LMArena Korean | 1339 | 1415 |
| LMArena Russian | 1377 | 1428 |
| LMArena Spanish | 1376 | 1448 |
Instruction Following GPT-5.6 Luna leads
GPT-4.1: 71.3 (#153), GPT-5.6 Luna: 75.6 (#57)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| LMArena Instruction Following | 1367 | 1437 |
| IFEval | 83.8% | — |
Long Context GPT-5.6 Luna leads
GPT-4.1: 40.0 (#163), GPT-5.6 Luna: 43.9 (#82)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| LMArena Longer Query | 1385 | 1436 |
| Fiction.LiveBench | 63.9% | — |
Writing & Preference GPT-5.6 Luna leads
GPT-4.1: 57.6 (#125), GPT-5.6 Luna: 68.0 (#29)
| Benchmark | GPT-4.1 | GPT-5.6 Luna |
|---|---|---|
| LMArena Text | 1383 | 1431 |
| LMArena Creative Writing | 1363 | 1396 |
| EQ-Bench Creative Writing | 1420 | 1829 |
| LMArena Multi-Turn | 1398 | 1434 |
| WildBench | 85.4% | — |
| EQ-Bench 4 | — | 1156 |
Frequently asked questions
Is GPT-4.1 better than GPT-5.6 Luna?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 35.9 on the Noometry Index.
Which is cheaper, GPT-4.1 or GPT-5.6 Luna?
GPT-5.6 Luna is cheaper. It lists at $0.20 per million input tokens and $1.20 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GPT-4.1 or GPT-5.6 Luna better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 1.05M.
How many benchmarks do GPT-4.1 and GPT-5.6 Luna share?
33 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and GPT-5.6 Luna has 52.