Model comparison
GPT-4.1 nano vs GPT-5.4
GPT-5.4 is the stronger model overall, scoring 59.5 to 27.7 on the Noometry Index. GPT-4.1 nano costs 32× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. GPT-4.1 nano scores higher in 0 categories and GPT-5.4 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.4 leads 61.7 to 8.6.
- The biggest single-benchmark swing is ARC-AGI-1: 0% for GPT-4.1 nano and 93.7% for GPT-5.4.
- GPT-4.1 nano is cheaper at $0.10 / $0.40 per million input/output tokens, against $2.50 / $15 for GPT-5.4.
- GPT-5.4 accepts more context: 1.05M tokens versus 1.05M.
Side by side
| GPT-4.1 nano | GPT-5.4 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 27.7 | 59.5 |
| Released | 2025-04-14 | 2026-03-05 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.05M |
| Max output | 33K | 128K |
| Input $ / M tokens | $0.10 | $2.50 |
| Output $ / M tokens | $0.40 | $15 |
| Results tracked | 38 | 68 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 leads
GPT-4.1 nano: 23.8 (#325), GPT-5.4: 52.8 (#28)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| SciCode | 25.9% | 56.6% |
| WeirdML | 19% | 77.7% |
| LMArena Coding | 1306 | 1497 |
| SWE-bench Verified | — | 76.9% |
| DeepSWE | — | 51.8% |
| Aider Polyglot | 8.9% | — |
| LMArena WebDev | — | 1465 |
| GSO | — | 31.4% |
| MirrorCode | — | 15.6% |
| ALE-Bench | — | 1,607 |
| AlgoTune | — | 1.85 |
Agentic & Tool Use GPT-5.4 leads
GPT-4.1 nano: 26.3 (#98), GPT-5.4: 46.5 (#12)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| Terminal-Bench | — | 81.8% |
| APEX-Agents | — | 52.4% |
| Berkeley Function Calling Leaderboard | 33% | — |
| τ²-bench Banking | — | 39.4% |
| DeepResearch Bench | — | 35.1% |
| PostTrainBench | — | 19% |
| GBAEval | — | 45.1% |
| LMArena Search | — | 1197 |
| METR Time Horizons | — | 74.3% |
| Vending-Bench 2 | — | 6,144 |
Reasoning GPT-5.4 leads
GPT-4.1 nano: 8.6 (#344), GPT-5.4: 61.7 (#17)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| ARC-AGI-2 | 0% | 74% |
| Kagi LLM Benchmark | 33.3% | 63.8% |
| ARC-AGI-1 | 0% | 93.7% |
| CritPt | 0% | 23.4% |
| LMArena Hard Prompts | 1286 | 1485 |
| DTBench | 52.5% | 94.4% |
| LMCA | 5.5% | 52% |
| Epoch Capabilities Index | 129.62 | 156.92 |
| NYT Connections (extended) | — | 91.3% |
| Chess Puzzles | — | 44% |
| EnigmaEval | — | 16% |
| Thematic Generalization | — | 80% |
| EBR-Bench | — | 25.4% |
| Mystery Game Puzzles | — | 37% |
| ForecastBench | — | 59.5 |
Math GPT-5.4 leads
GPT-4.1 nano: 25.5 (#246), GPT-5.4: 73.3 (#14)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 28.9% | 97.8% |
| LMArena Math | 1274 | 1488 |
| FrontierMath (Feb 2025 set) | 1% | 47.6% |
| FrontierMath (Tiers 1-3) | — | 78.6% |
| FrontierMath Tier 4 | — | 49% |
| MathArena Final-Answer Competitions | — | 83.1% |
| ProofBench | — | 56% |
| Omni-MATH | 36.7% | — |
| MATH Level 5 | 70% | — |
| FrontierMath Tier 4 (v1) | — | 27.1% |
Knowledge GPT-5.4 leads
GPT-4.1 nano: 21.8 (#268), GPT-5.4: 65.3 (#13)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| GPQA Diamond | 48.9% | 93.3% |
| SimpleQA Verified | 6% | 45.1% |
| LMArena Expert | 1272 | 1507 |
| Humanity's Last Exam | — | 36.2% |
| MMLU-Pro | 55% | — |
| Vectara Hallucination Rate | — | 7% |
| GPQA (HELM) | 50.7% | — |
Multimodal GPT-5.4 leads
GPT-4.1 nano: 28.6 (#108), GPT-5.4: 43.8 (#19)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| LMArena Vision | 1063 | 1302 |
| Blueprint-Bench 2 | — | 27.1% |
| Furniture Assembly | — | 37.5% |
| LMArena Document | — | 1471 |
Multilingual GPT-5.4 leads
GPT-4.1 nano: 41.8 (#200), GPT-5.4: 56.4 (#22)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| LMArena Non-English | 1260 | 1465 |
| LMArena Chinese | 1270 | 1520 |
| LMArena German | 1288 | 1472 |
| LMArena Japanese | 1199 | 1484 |
| LMArena Russian | 1261 | 1480 |
| LMArena French | — | 1493 |
| LMArena Korean | — | 1449 |
| LMArena Spanish | — | 1453 |
Instruction Following GPT-5.4 leads
GPT-4.1 nano: 68.0 (#188), GPT-5.4: 77.3 (#25)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| LMArena Instruction Following | 1267 | 1469 |
| IFEval | 84.3% | — |
Long Context GPT-5.4 leads
GPT-4.1 nano: 23.7 (#291), GPT-5.4: 50.3 (#8)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| LMArena Longer Query | 1283 | 1474 |
| Fiction.LiveBench | 25% | — |
| CL-bench | — | 27.9% |
| CL-bench Life | — | 21.7% |
Writing & Preference GPT-5.4 leads
GPT-4.1 nano: 40.7 (#238), GPT-5.4: 72.1 (#18)
| Benchmark | GPT-4.1 nano | GPT-5.4 |
|---|---|---|
| LMArena Text | 1285 | 1469 |
| LMArena Creative Writing | 1260 | 1439 |
| EQ-Bench Creative Writing | 946 | 1840 |
| LMArena Multi-Turn | 1277 | 1482 |
| WildBench | 81.2% | — |
| EQ-Bench 4 | — | 1272 |
Frequently asked questions
Is GPT-4.1 nano better than GPT-5.4?
GPT-5.4 is the stronger model overall, scoring 59.5 to 27.7 on the Noometry Index. GPT-4.1 nano costs 32× less per token, which makes it the better buy when GPT-5.4's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 nano or GPT-5.4?
GPT-4.1 nano is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; GPT-5.4 lists at $2.50 and $15.
Is GPT-4.1 nano or GPT-5.4 better for coding?
GPT-5.4 scores higher on coding benchmarks: 52.8 versus 23.8 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 does, with 1.05M tokens against 1.05M.
How many benchmarks do GPT-4.1 nano and GPT-5.4 share?
29 benchmarks have published results for both models. GPT-4.1 nano has 38 scored results on Noometry and GPT-5.4 has 68.