Model comparison
GPT-5.4 nano vs o1
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 40.9 on the Noometry Index.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. GPT-5.4 nano scores higher in 5 categories and o1 in 4 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in long context, where o1 leads 50.3 to 41.6.
- The biggest single-benchmark swing is FrontierMath (Tiers 1-3): 44.9% for GPT-5.4 nano and 14.7% for o1.
- GPT-5.4 nano is cheaper at $0.20 / $1.25 per million input/output tokens, against $15 / $60 for o1.
- GPT-5.4 nano accepts more context: 400K tokens versus 200K.
Side by side
| GPT-5.4 nano | o1 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 41.9 | 40.9 |
| Released | 2026-03-17 | 2024-09-12 |
| Weights | Proprietary | Proprietary |
| Context window | 400K | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $0.20 | $15 |
| Output $ / M tokens | $1.25 | $60 |
| Results tracked | 40 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o1 leads
GPT-5.4 nano: 43.6 (#84), o1: 46.1 (#70)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| WeirdML | 49.2% | 47.6% |
| LMArena Coding | 1405 | 1367 |
| Aider Polyglot | — | 61.7% |
| SciCode | 46.9% | — |
| LiveBench Coding | — | 69.7% |
| CadEval | — | 56% |
| ALE-Bench | 1,005 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 80.2% |
Agentic & Tool Use Not comparable
GPT-5.4 nano: —, o1: 24.6 (#117)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| Cybench | — | 10% |
| METR Time Horizons | — | 51.1% |
Reasoning o1 leads
GPT-5.4 nano: 23.7 (#173), o1: 27.9 (#111)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| ARC-AGI-1 | 51.5% | 30.7% |
| Chess Puzzles | 30% | 15% |
| LMArena Hard Prompts | 1381 | 1371 |
| DTBench | 80.3% | 74.7% |
| LMCA | 36.9% | 22.3% |
| Epoch Capabilities Index | 145.81 | 141.91 |
| ARC-AGI-2 | 5.7% | — |
| SimpleBench | — | 41.7% |
| Kagi LLM Benchmark | 39.7% | — |
| CritPt | 9.3% | — |
| EnigmaEval | — | 5.7% |
| LiveBench Reasoning | — | 91.6% |
| Mystery Game Puzzles | 9% | — |
| LiveBench Data Analysis | — | 65.5% |
| ForecastBench | 57.3 | — |
| LiveBench | — | 75.7% |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.9 (#88), o1: 36.1 (#175)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 44.9% | 14.7% |
| OTIS Mock AIME 2024-2025 | 87.8% | 73.3% |
| LMArena Math | 1406 | 1388 |
| FrontierMath (Feb 2025 set) | 25.9% | 9.3% |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 5% | — |
| LiveBench Math | — | 80.3% |
| MATH Level 5 | — | 94.7% |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge Too close to call
GPT-5.4 nano: 41.9 (#103), o1: 41.5 (#110)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| GPQA Diamond | 78.5% | 76.8% |
| SimpleQA Verified | 11.7% | 41.1% |
| LMArena Expert | 1396 | 1361 |
| Humanity's Last Exam | — | 8% |
| Confabulations | — | 11.7% |
| Vectara Hallucination Rate | 3.1% | — |
Multimodal GPT-5.4 nano leads
GPT-5.4 nano: 36.7 (#78), o1: 34.2 (#93)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| LMArena Vision | 1196 | 1168 |
| GeoBench | — | 80% |
| VPCT | — | 37% |
| SpatialViz-Bench | — | 41.4% |
Multilingual Too close to call
GPT-5.4 nano: 48.6 (#140), o1: 48.6 (#142)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| LMArena Non-English | 1359 | 1358 |
| LMArena Chinese | 1392 | 1394 |
| LMArena French | 1396 | 1344 |
| LMArena German | 1367 | 1337 |
| LMArena Japanese | 1343 | 1346 |
| LMArena Korean | 1320 | 1396 |
| LMArena Russian | 1363 | 1356 |
| LMArena Spanish | 1371 | 1345 |
Instruction Following o1 leads
GPT-5.4 nano: 71.9 (#144), o1: 74.8 (#86)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| LMArena Instruction Following | 1362 | 1367 |
| LiveBench Instruction Following | — | 81.5% |
Long Context o1 leads
GPT-5.4 nano: 41.6 (#137), o1: 50.3 (#9)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| LMArena Longer Query | 1366 | 1378 |
| Fiction.LiveBench | — | 83.3% |
Writing & Preference Too close to call
GPT-5.4 nano: 55.7 (#142), o1: 55.6 (#144)
| Benchmark | GPT-5.4 nano | o1 |
|---|---|---|
| LMArena Text | 1372 | 1366 |
| LMArena Creative Writing | 1314 | 1348 |
| LMArena Multi-Turn | 1382 | 1369 |
| Short-Story Creative Writing | — | 70.2% |
| LiveBench Language | — | 65.4% |
Frequently asked questions
Is GPT-5.4 nano better than o1?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 40.9 on the Noometry Index.
Which is cheaper, GPT-5.4 nano or o1?
GPT-5.4 nano is cheaper. It lists at $0.20 per million input tokens and $1.25 per million output tokens; o1 lists at $15 and $60.
Is GPT-5.4 nano or o1 better for coding?
o1 scores higher on coding benchmarks: 46.1 versus 43.6 in the Noometry coding category.
Which has the bigger context window?
GPT-5.4 nano does, with 400K tokens against 200K.
How many benchmarks do GPT-5.4 nano and o1 share?
29 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and o1 has 52.