Model comparison
GPT-5.1 vs GPT-5.4 nano
GPT-5.1 is the stronger model overall, scoring 49.0 to 41.9 on the Noometry Index. GPT-5.4 nano costs 7.4× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Last verified . 36 shared benchmarks.
Summary
- They share 36 benchmarks with published results for both. GPT-5.1 scores higher in 9 categories and GPT-5.4 nano in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.1 leads 39.8 to 23.7.
- The biggest single-benchmark swing is SimpleQA Verified: 48% for GPT-5.1 and 11.7% for GPT-5.4 nano.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
Side by side
| GPT-5.1 | GPT-5.4 nano | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 49.0 | 41.9 |
| Released | 2025-11-13 | 2026-03-17 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $1.25 | $0.20 |
| Output $ / M tokens | $10 | $1.25 |
| Results tracked | 63 | 40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
GPT-5.1: 46.4 (#66), GPT-5.4 nano: 43.6 (#84)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| SciCode | 43.3% | 46.9% |
| WeirdML | 60.8% | 49.2% |
| LMArena Coding | 1454 | 1405 |
| ALE-Bench | 1,192 | 1,005 |
| SWE-bench Verified | 68% | — |
| SWE-bench Verified (bash only) | 66% | — |
| LMArena WebDev | 1395 | — |
| GSO | 13.7% | — |
| LiveBench Coding | 72.5% | — |
Agentic & Tool Use Not comparable
GPT-5.1: 32.7 (#60), GPT-5.4 nano: —
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| Terminal-Bench | 47.6% | — |
| DeepResearch Bench | 42.8% | — |
| LMArena Search | 1199 | — |
| Vending-Bench 2 | 1,473 | — |
Reasoning GPT-5.1 leads
GPT-5.1: 39.8 (#58), GPT-5.4 nano: 23.7 (#173)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| ARC-AGI-2 | 17.6% | 5.7% |
| ARC-AGI-1 | 72.8% | 51.5% |
| CritPt | 4.9% | 9.3% |
| Chess Puzzles | 32% | 30% |
| LMArena Hard Prompts | 1457 | 1381 |
| Mystery Game Puzzles | 19% | 9% |
| DTBench | 90.1% | 80.3% |
| LMCA | 43.9% | 36.9% |
| Epoch Capabilities Index | 149.64 | 145.81 |
| ForecastBench | 58.1 | 57.3 |
| SimpleBench | 53.2% | — |
| Kagi LLM Benchmark | — | 39.7% |
| EnigmaEval | 11.2% | — |
| LiveBench Reasoning | 95.8% | — |
| LiveBench Data Analysis | 72.1% | — |
| LiveBench | 78.8% | — |
Math GPT-5.1 leads
GPT-5.1: 52.2 (#51), GPT-5.4 nano: 40.9 (#88)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 87.8% |
| LMArena Math | 1447 | 1406 |
| FrontierMath (Feb 2025 set) | 31% | 25.9% |
| FrontierMath Tier 4 (v1) | 12.5% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 44.9% |
| FrontierMath Tier 4 | — | 12.2% |
| ProofBench | — | 5% |
| Omni-MATH | 46.4% | — |
| LiveBench Math | 94.5% | — |
Knowledge GPT-5.1 leads
GPT-5.1: 50.6 (#71), GPT-5.4 nano: 41.9 (#103)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| GPQA Diamond | 87.6% | 78.5% |
| SimpleQA Verified | 48% | 11.7% |
| Vectara Hallucination Rate | 10.9% | 3.1% |
| LMArena Expert | 1470 | 1396 |
| Humanity's Last Exam | 23.7% | — |
| MMLU-Pro | 57.9% | — |
| GPQA (HELM) | 44.2% | — |
Multimodal GPT-5.1 leads
GPT-5.1: 44.8 (#19), GPT-5.4 nano: 36.7 (#78)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| LMArena Vision | 1250 | 1196 |
| VPCT | 58.7% | — |
| LMArena Document | 1403 | — |
Multilingual GPT-5.1 leads
GPT-5.1: 53.8 (#56), GPT-5.4 nano: 48.6 (#140)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| LMArena Non-English | 1431 | 1359 |
| LMArena Chinese | 1495 | 1392 |
| LMArena French | 1450 | 1396 |
| LMArena German | 1438 | 1367 |
| LMArena Japanese | 1453 | 1343 |
| LMArena Korean | 1401 | 1320 |
| LMArena Russian | 1435 | 1363 |
| LMArena Spanish | 1433 | 1371 |
Instruction Following GPT-5.1 leads
GPT-5.1: 83.9 (#1), GPT-5.4 nano: 71.9 (#144)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| LMArena Instruction Following | 1443 | 1362 |
| LiveBench Instruction Following | 93.3% | — |
| IFEval | 93.5% | — |
Long Context GPT-5.1 leads
GPT-5.1: 47.6 (#14), GPT-5.4 nano: 41.6 (#137)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| LMArena Longer Query | 1447 | 1366 |
| CL-bench | 23.7% | — |
| CL-bench Life | 17.3% | — |
Writing & Preference GPT-5.1 leads
GPT-5.1: 64.5 (#55), GPT-5.4 nano: 55.7 (#142)
| Benchmark | GPT-5.1 | GPT-5.4 nano |
|---|---|---|
| LMArena Text | 1443 | 1372 |
| LMArena Creative Writing | 1427 | 1314 |
| LMArena Multi-Turn | 1450 | 1382 |
| WildBench | 86.3% | — |
| LiveBench Language | 80.2% | — |
Frequently asked questions
Is GPT-5.1 better than GPT-5.4 nano?
GPT-5.1 is the stronger model overall, scoring 49.0 to 41.9 on the Noometry Index. GPT-5.4 nano costs 7.4× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Which is cheaper, GPT-5.1 or GPT-5.4 nano?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; GPT-5.1 lists at $1.25 and $10.
Is GPT-5.1 or GPT-5.4 nano better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 43.6 in the Noometry coding category.
Which has the bigger context window?
Both accept 400K tokens.
How many benchmarks do GPT-5.1 and GPT-5.4 nano share?
36 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and GPT-5.4 nano has 40.