Model comparison
GPT-4.1 vs GPT-5 Nano
GPT-4.1 is the stronger model overall, scoring 35.9 to 33.5 on the Noometry Index. GPT-5 Nano costs 25× less per token, which makes it the better buy when GPT-4.1's lead doesn't matter for your workload.
Last verified . 44 shared benchmarks.
Summary
- They share 44 benchmarks with published results for both. GPT-4.1 scores higher in 7 categories and GPT-5 Nano in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.1 leads 57.6 to 39.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 38.3% for GPT-4.1 and 81.1% for GPT-5 Nano.
- GPT-5 Nano is cheaper at $0.05 / $0.40 per million input/output tokens, against $2 / $8 for GPT-4.1.
- GPT-4.1 accepts more context: 1.05M tokens versus 400K.
Side by side
| GPT-4.1 | GPT-5 Nano | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 35.9 | 33.5 |
| Released | 2025-04-14 | 2025-08-07 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 400K |
| Max output | 33K | 128K |
| Input $ / M tokens | $2 | $0.05 |
| Output $ / M tokens | $8 | $0.40 |
| Results tracked | 52 | 49 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-4.1: 34.4 (#238), GPT-5 Nano: 33.6 (#254)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| SWE-bench Verified (bash only) | 39.6% | 34.8% |
| WeirdML | 39% | 38.1% |
| LMArena Coding | 1391 | 1351 |
| ALE-Bench | 558.1 | 718.67 |
| SWE-bench Verified | 48.5% | — |
| Aider Polyglot | 52.4% | — |
| CadEval | 42% | — |
Agentic & Tool Use GPT-4.1 leads
GPT-4.1: 34.7 (#43), GPT-5 Nano: 25.8 (#106)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| Berkeley Function Calling Leaderboard | 54% | 51.5% |
| Terminal-Bench | — | 21.8% |
Reasoning GPT-5 Nano leads
GPT-4.1: 11.7 (#339), GPT-5 Nano: 16.3 (#306)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| ARC-AGI-2 | 0.4% | 2.6% |
| Kagi LLM Benchmark | 52.3% | 62.2% |
| ARC-AGI-1 | 5.5% | 20.7% |
| Chess Puzzles | 6% | 27% |
| LMArena Hard Prompts | 1384 | 1328 |
| DTBench | 68.3% | 62.7% |
| LMCA | 25.6% | 7.9% |
| Epoch Capabilities Index | 136.78 | 139.38 |
| ForecastBench | 61.5 | 59.1 |
| SimpleBench | 27% | — |
| EnigmaEval | 2.2% | — |
| Mystery Game Puzzles | — | 9% |
Math GPT-5 Nano leads
GPT-4.1: 22.3 (#280), GPT-5 Nano: 29.4 (#241)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| FrontierMath (Tiers 1-3) | 6% | 20% |
| OTIS Mock AIME 2024-2025 | 38.3% | 81.1% |
| Omni-MATH | 47.1% | 54.6% |
| LMArena Math | 1370 | 1317 |
| MATH Level 5 | 83% | 95.2% |
| FrontierMath (Feb 2025 set) | 5.5% | 8.3% |
| FrontierMath Tier 4 (v1) | 0% | 2.1% |
| FrontierMath Tier 4 | — | 2.4% |
| ProofBench | — | 12% |
Knowledge GPT-4.1 leads
GPT-4.1: 37.1 (#160), GPT-5 Nano: 35.9 (#178)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| GPQA Diamond | 66.9% | 69.4% |
| SimpleQA Verified | 31.1% | 11.7% |
| MMLU-Pro | 81.1% | 77.8% |
| Vectara Hallucination Rate | 5.6% | 10.5% |
| GPQA (HELM) | 65.9% | 67.9% |
| LMArena Expert | 1364 | 1321 |
| Humanity's Last Exam | 5.4% | — |
Multimodal GPT-4.1 leads
GPT-4.1: 38.2 (#67), GPT-5 Nano: 31.3 (#108)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| LMArena Vision | 1211 | 1159 |
| GeoBench | 72% | — |
| VPCT | — | 37.2% |
Multilingual GPT-4.1 leads
GPT-4.1: 49.4 (#133), GPT-5 Nano: 45.3 (#172)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | 1370 | 1313 |
| LMArena Chinese | 1382 | 1356 |
| LMArena German | 1381 | 1327 |
| LMArena Japanese | 1319 | 1226 |
| LMArena Korean | 1339 | 1269 |
| LMArena Russian | 1377 | 1296 |
| LMArena Spanish | 1376 | 1360 |
| LMArena French | 1382 | — |
Instruction Following GPT-5 Nano leads
GPT-4.1: 71.3 (#153), GPT-5 Nano: 75.0 (#79)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| IFEval | 83.8% | 93.2% |
| LMArena Instruction Following | 1367 | 1306 |
Long Context GPT-4.1 leads
GPT-4.1: 40.0 (#163), GPT-5 Nano: 31.3 (#281)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| Fiction.LiveBench | 63.9% | 44.4% |
| LMArena Longer Query | 1385 | 1312 |
Writing & Preference GPT-4.1 leads
GPT-4.1: 57.6 (#125), GPT-5 Nano: 39.1 (#249)
| Benchmark | GPT-4.1 | GPT-5 Nano |
|---|---|---|
| LMArena Text | 1383 | 1320 |
| LMArena Creative Writing | 1363 | 1249 |
| EQ-Bench Creative Writing | 1420 | 705 |
| WildBench | 85.4% | 80.6% |
| LMArena Multi-Turn | 1398 | 1311 |
Frequently asked questions
Is GPT-4.1 better than GPT-5 Nano?
GPT-4.1 is the stronger model overall, scoring 35.9 to 33.5 on the Noometry Index. GPT-5 Nano costs 25× 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-5 Nano?
GPT-5 Nano is cheaper. It lists at $0.05 per million input tokens and $0.40 per million output tokens; GPT-4.1 lists at $2 and $8.
Is GPT-4.1 or GPT-5 Nano better for coding?
They score almost the same on coding (34.4 vs 33.6); test both on your own repository before choosing.
Which has the bigger context window?
GPT-4.1 does, with 1.05M tokens against 400K.
How many benchmarks do GPT-4.1 and GPT-5 Nano share?
44 benchmarks have published results for both models. GPT-4.1 has 52 scored results on Noometry and GPT-5 Nano has 49.