Model comparison
GPT-4.1 vs GPT-5.6 Sol
GPT-5.6 Sol is the stronger model overall, scoring 64.4 to 35.7 on the Noometry Index. GPT-4.1 costs 2.3× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.
Last verified . 35 shared benchmarks.
Summary
- They share 35 benchmarks with published results for both. GPT-4.1 scores higher in 0 categories and GPT-5.6 Sol in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Sol leads 85.5 to 21.4.
- The biggest single-benchmark swing is ARC-AGI-2: 0.4% for GPT-4.1 and 92.5% for GPT-5.6 Sol.
- GPT-4.1 is cheaper at $2 / $8 per million input/output tokens, against $4 / $20 for GPT-5.6 Sol.
- GPT-5.6 Sol accepts more context: 1.05M tokens versus 1.05M.
Side by side
| GPT-4.1 | GPT-5.6 Sol | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.7 | 64.4 |
| Released | 2025-04-14 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 33K | 128K |
| Input $ / M tokens | $2 | $4 |
| Output $ / M tokens | $8 | $20 |
| Results tracked | 52 | 64 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Sol leads
GPT-4.1: 34.2 (#228), GPT-5.6 Sol: 61.1 (#7)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| WeirdML | 39% | 89.4% |
| LMArena Coding | 1391 | 1493 |
| ALE-Bench | 558.1 | 2,177 |
| SWE-bench Verified | 48.5% | — |
| DeepSWE | — | 72.7% |
| FrontierCode | — | 47.5% |
| SWE-bench Verified (bash only) | 39.6% | — |
| Aider Polyglot | 52.4% | — |
| CursorBench | — | 41.7% |
| LMArena WebDev | — | 1619 |
| FrontierSWE | — | 32.2% |
| SciCode | — | 57.1% |
| MirrorCode | — | 20% |
| CadEval | 42% | — |
Agentic & Tool Use GPT-5.6 Sol leads
GPT-4.1: 34.4 (#40), GPT-5.6 Sol: 50.2 (#7)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| APEX-Agents | — | 51.4% |
| Berkeley Function Calling Leaderboard | 54% | — |
| OSWorld 2.0 | — | 27.3% |
| τ²-bench Banking | — | 46.9% |
| PostTrainBench | — | 36.2% |
| BALROG | — | 60% |
| GBAEval | — | 52.6% |
| GDP.pdf | — | 30.7% |
| LMArena Search | — | 1257 |
| Vending-Bench 2 | — | 9,619 |
Reasoning GPT-5.6 Sol leads
GPT-4.1: 11.7 (#334), GPT-5.6 Sol: 74.7 (#6)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| ARC-AGI-2 | 0.4% | 92.5% |
| SimpleBench | 27% | 71.7% |
| Kagi LLM Benchmark | 52.3% | 67% |
| ARC-AGI-1 | 5.5% | 97.5% |
| Chess Puzzles | 6% | 64% |
| EnigmaEval | 2.2% | 37.1% |
| LMArena Hard Prompts | 1384 | 1481 |
| DTBench | 68.3% | 96% |
| LMCA | 25.6% | 59.2% |
| Epoch Capabilities Index | 136.79 | 161.99 |
| NYT Connections (extended) | — | 93.8% |
| CritPt | — | 32.3% |
| EBR-Bench | — | 44.8% |
| Mystery Game Puzzles | — | 58% |
| Surface Evolver Bench | — | 93.1% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | 61.5 | — |
Math GPT-5.6 Sol leads
GPT-4.1: 21.4 (#269), GPT-5.6 Sol: 85.5 (#7)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 89.1% |
| OTIS Mock AIME 2024-2025 | 38.3% | 100% |
| LMArena Math | 1369 | 1472 |
| FrontierMath Tier 4 | — | 82.9% |
| ProofBench | — | 83% |
| Omni-MATH | 47.1% | — |
| MATH Level 5 | 83% | — |
| FrontierMath (Feb 2025 set) | 5.5% | — |
| FrontierMath Erdős | — | 0% |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-5.6 Sol leads
GPT-4.1: 37.0 (#152), GPT-5.6 Sol: 64.3 (#16)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| GPQA Diamond | 66.9% | 93.5% |
| SimpleQA Verified | 31.1% | 69.7% |
| Vectara Hallucination Rate | 5.6% | 12.4% |
| LMArena Expert | 1364 | 1517 |
| Humanity's Last Exam | 5.4% | — |
| MMLU-Pro | 81.1% | — |
| GPQA (HELM) | 65.9% | — |
Multimodal GPT-5.6 Sol leads
GPT-4.1: 37.5 (#66), GPT-5.6 Sol: 48.9 (#8)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| LMArena Vision | 1210 | 1281 |
| GeoBench | 72% | — |
| Blueprint-Bench 2 | — | 33.6% |
| Furniture Assembly | — | 56.7% |
| LMArena Document | — | 1483 |
Multilingual GPT-5.6 Sol leads
GPT-4.1: 49.6 (#128), GPT-5.6 Sol: 55.4 (#32)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| LMArena Non-English | 1370 | 1450 |
| LMArena Chinese | 1382 | 1523 |
| LMArena French | 1383 | 1475 |
| LMArena German | 1380 | 1476 |
| LMArena Japanese | 1320 | 1475 |
| LMArena Korean | 1339 | 1434 |
| LMArena Russian | 1377 | 1466 |
| LMArena Spanish | 1375 | 1449 |
Instruction Following GPT-5.6 Sol leads
GPT-4.1: 71.5 (#148), GPT-5.6 Sol: 77.8 (#16)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| LMArena Instruction Following | 1367 | 1480 |
| IFEval | 83.8% | — |
Long Context GPT-5.6 Sol leads
GPT-4.1: 40.0 (#158), GPT-5.6 Sol: 45.3 (#42)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| LMArena Longer Query | 1385 | 1479 |
| Fiction.LiveBench | 63.9% | — |
Writing & Preference GPT-5.6 Sol leads
GPT-4.1: 57.8 (#120), GPT-5.6 Sol: 73.5 (#12)
| Benchmark | GPT-4.1 | GPT-5.6 Sol |
|---|---|---|
| LMArena Text | 1383 | 1456 |
| LMArena Creative Writing | 1363 | 1450 |
| EQ-Bench Creative Writing | 1420 | 1972 |
| LMArena Multi-Turn | 1398 | 1456 |
| WildBench | 85.4% | — |
| EQ-Bench 4 | — | 1250 |
Frequently asked questions
Is GPT-4.1 better than GPT-5.6 Sol?
GPT-5.6 Sol is the stronger model overall, scoring 64.4 to 35.7 on the Noometry Index. GPT-4.1 costs 2.3× less per token, which makes it the better buy when GPT-5.6 Sol's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 or GPT-5.6 Sol?
GPT-4.1 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; GPT-5.6 Sol lists at $4 and $20.
Is GPT-4.1 or GPT-5.6 Sol better for coding?
GPT-5.6 Sol scores higher on coding benchmarks: 61.1 versus 34.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Sol does, with 1.05M tokens against 1.05M.
How many benchmarks do GPT-4.1 and GPT-5.6 Sol share?
35 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and GPT-5.6 Sol has 64.