Model comparison
GPT-4.1 vs GPT-6 Sol
GPT-6 Sol is the stronger model overall, scoring 61.8 to 35.9 on the Noometry Index.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GPT-4.1 scores higher in 0 categories and GPT-6 Sol in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Sol leads 87.2 to 22.3.
- The biggest single-benchmark swing is ARC-AGI-1: 5.5% for GPT-4.1 and 95.5% for GPT-6 Sol.
- GPT-4.1 is cheaper at $2 / $8 per million input/output tokens, against $2 / $10 for GPT-6 Sol.
- GPT-6 Sol accepts more context: 1.05M tokens versus 1.05M.
Side by side
| GPT-4.1 | GPT-6 Sol | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.9 | 61.8 |
| Released | 2025-04-14 | 2026-09-22 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 33K | 128K |
| Input $ / M tokens | $2 | $2 |
| Output $ / M tokens | $8 | $10 |
| Results tracked | 52 | 45 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-6 Sol leads
GPT-4.1: 34.4 (#238), GPT-6 Sol: 60.1 (#11)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| LMArena Coding | 1391 | 1447 |
| ALE-Bench | 558.1 | 2,462 |
| SWE-bench Verified | 48.5% | — |
| DeepSWE | — | 68.8% |
| FrontierCode | — | 49.3% |
| SWE-bench Verified (bash only) | 39.6% | — |
| Aider Polyglot | 52.4% | — |
| LMArena WebDev | — | 1688 |
| SciCode | — | 57.6% |
| WeirdML | 39% | — |
| CadEval | 42% | — |
Agentic & Tool Use GPT-6 Sol leads
GPT-4.1: 34.7 (#43), GPT-6 Sol: 37.2 (#36)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| APEX-Agents | — | 54.3% |
| Berkeley Function Calling Leaderboard | 54% | — |
| GDP.pdf | — | 26.4% |
| Vending-Bench 2 | — | 14,428 |
Reasoning GPT-6 Sol leads
GPT-4.1: 11.7 (#339), GPT-6 Sol: 74.0 (#9)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| ARC-AGI-2 | 0.4% | 89.6% |
| ARC-AGI-1 | 5.5% | 95.5% |
| LMArena Hard Prompts | 1384 | 1418 |
| DTBench | 68.3% | 97.3% |
| LMCA | 25.6% | 59.1% |
| Epoch Capabilities Index | 136.78 | 162.72 |
| SimpleBench | 27% | — |
| Kagi LLM Benchmark | 52.3% | — |
| NYT Connections (extended) | — | 90.1% |
| CritPt | — | 30.9% |
| Chess Puzzles | 6% | — |
| EnigmaEval | 2.2% | — |
| EBR-Bench | — | 53.3% |
| Mystery Game Puzzles | — | 56% |
| ForecastBench | 61.5 | — |
Math GPT-6 Sol leads
GPT-4.1: 22.3 (#280), GPT-6 Sol: 87.2 (#7)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 89.8% |
| OTIS Mock AIME 2024-2025 | 38.3% | 100% |
| LMArena Math | 1370 | 1402 |
| FrontierMath Tier 4 | — | 90% |
| ProofBench | — | 83% |
| Omni-MATH | 47.1% | — |
| MATH Level 5 | 83% | — |
| FrontierMath (Feb 2025 set) | 5.5% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-6 Sol leads
GPT-4.1: 37.1 (#160), GPT-6 Sol: 64.8 (#15)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| GPQA Diamond | 66.9% | 94.3% |
| SimpleQA Verified | 31.1% | 60.7% |
| Vectara Hallucination Rate | 5.6% | 6.5% |
| LMArena Expert | 1364 | 1439 |
| Humanity's Last Exam | 5.4% | — |
| MMLU-Pro | 81.1% | — |
| GPQA (HELM) | 65.9% | — |
Multimodal GPT-6 Sol leads
GPT-4.1: 38.2 (#67), GPT-6 Sol: 47.6 (#10)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| LMArena Vision | 1211 | 1245 |
| GeoBench | 72% | — |
| Blueprint-Bench 2 | — | 36.9% |
| Furniture Assembly | — | 58.3% |
Multilingual GPT-6 Sol leads
GPT-4.1: 49.4 (#133), GPT-6 Sol: 50.5 (#118)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| LMArena Non-English | 1370 | 1385 |
| LMArena Chinese | 1382 | 1405 |
| LMArena French | 1382 | 1410 |
| LMArena German | 1381 | 1390 |
| LMArena Japanese | 1319 | 1385 |
| LMArena Korean | 1339 | 1341 |
| LMArena Russian | 1377 | 1401 |
| LMArena Spanish | 1376 | 1384 |
Instruction Following GPT-6 Sol leads
GPT-4.1: 71.3 (#153), GPT-6 Sol: 74.5 (#94)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| LMArena Instruction Following | 1367 | 1412 |
| IFEval | 83.8% | — |
Long Context GPT-6 Sol leads
GPT-4.1: 40.0 (#163), GPT-6 Sol: 43.1 (#108)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| LMArena Longer Query | 1385 | 1411 |
| Fiction.LiveBench | 63.9% | — |
Writing & Preference GPT-6 Sol leads
GPT-4.1: 57.6 (#125), GPT-6 Sol: 71.9 (#18)
| Benchmark | GPT-4.1 | GPT-6 Sol |
|---|---|---|
| LMArena Text | 1383 | 1395 |
| LMArena Creative Writing | 1363 | 1378 |
| EQ-Bench Creative Writing | 1420 | 2125 |
| LMArena Multi-Turn | 1398 | 1412 |
| WildBench | 85.4% | — |
Frequently asked questions
Is GPT-4.1 better than GPT-6 Sol?
GPT-6 Sol is the stronger model overall, scoring 61.8 to 35.9 on the Noometry Index.
Which is cheaper, GPT-4.1 or GPT-6 Sol?
GPT-4.1 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; GPT-6 Sol lists at $2 and $10.
Is GPT-4.1 or GPT-6 Sol better for coding?
GPT-6 Sol scores higher on coding benchmarks: 60.1 versus 34.4 in the Noometry coding category.
Which has the bigger context window?
GPT-6 Sol does, with 1.05M tokens against 1.05M.
How many benchmarks do GPT-4.1 and GPT-6 Sol share?
30 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and GPT-6 Sol has 45.