Model comparison
GPT-4.1 nano vs o3
o3 is the stronger model overall, scoring 47.4 to 27.7 on the Noometry Index. GPT-4.1 nano costs 20× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Last verified . 37 shared benchmarks.
Summary
- They share 37 benchmarks with published results for both. GPT-4.1 nano scores higher in 0 categories and o3 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where o3 leads 54.5 to 21.8.
- The biggest single-benchmark swing is Aider Polyglot: 8.9% for GPT-4.1 nano and 81.3% for o3.
- GPT-4.1 nano is cheaper at $0.10 / $0.40 per million input/output tokens, against $2 / $8 for o3.
- GPT-4.1 nano accepts more context: 1.05M tokens versus 200K.
Side by side
| GPT-4.1 nano | o3 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 27.7 | 47.4 |
| Released | 2025-04-14 | 2025-04-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 33K | 100K |
| Input $ / M tokens | $0.10 | $2 |
| Output $ / M tokens | $0.40 | $8 |
| Results tracked | 38 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3 leads
GPT-4.1 nano: 23.8 (#325), o3: 47.3 (#57)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| Aider Polyglot | 8.9% | 81.3% |
| WeirdML | 19% | 52.4% |
| LMArena Coding | 1306 | 1408 |
| SWE-bench Verified | — | 62.3% |
| SWE-bench Verified (bash only) | — | 58.4% |
| SciCode | 25.9% | — |
| GSO | — | 8.8% |
| CadEval | — | 74% |
| ALE-Bench | — | 933.55 |
Agentic & Tool Use o3 leads
GPT-4.1 nano: 26.3 (#98), o3: 34.3 (#42)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 33% | 63% |
| GDPval | — | 30.8% |
| DeepResearch Bench | — | 45.2% |
| OSWorld | — | 23% |
| LMArena Search | — | 1144 |
| METR Time Horizons | — | 65.4% |
Reasoning o3 leads
GPT-4.1 nano: 8.6 (#344), o3: 32.1 (#66)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| ARC-AGI-2 | 0% | 6.5% |
| Kagi LLM Benchmark | 33.3% | 67.6% |
| ARC-AGI-1 | 0% | 60.8% |
| CritPt | 0% | 1.4% |
| LMArena Hard Prompts | 1286 | 1402 |
| DTBench | 52.5% | 84.8% |
| LMCA | 5.5% | 39.7% |
| Epoch Capabilities Index | 129.62 | 146.88 |
| SimpleBench | — | 53.1% |
| Chess Puzzles | — | 38% |
| EnigmaEval | — | 13.1% |
| Mystery Game Puzzles | — | 29% |
| ForecastBench | — | 62.5 |
Math o3 leads
GPT-4.1 nano: 25.5 (#246), o3: 48.9 (#49)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 28.9% | 84.4% |
| Omni-MATH | 36.7% | 71.4% |
| LMArena Math | 1274 | 1426 |
| MATH Level 5 | 70% | 97.8% |
| FrontierMath (Feb 2025 set) | 1% | 18.7% |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge o3 leads
GPT-4.1 nano: 21.8 (#268), o3: 54.5 (#47)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| GPQA Diamond | 48.9% | 81.8% |
| SimpleQA Verified | 6% | 49.4% |
| MMLU-Pro | 55% | 85.9% |
| GPQA (HELM) | 50.7% | 75.3% |
| LMArena Expert | 1272 | 1403 |
| Humanity's Last Exam | — | 20.3% |
| Confabulations | — | 14.4% |
Multimodal o3 leads
GPT-4.1 nano: 28.6 (#108), o3: 40.7 (#35)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| LMArena Vision | 1063 | 1214 |
| GeoBench | — | 74% |
| VPCT | — | 52% |
Multilingual o3 leads
GPT-4.1 nano: 41.8 (#200), o3: 51.9 (#99)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| LMArena Non-English | 1260 | 1402 |
| LMArena Chinese | 1270 | 1438 |
| LMArena German | 1288 | 1420 |
| LMArena Japanese | 1199 | 1404 |
| LMArena Russian | 1261 | 1407 |
| LMArena French | — | 1430 |
| LMArena Korean | — | 1371 |
| LMArena Spanish | — | 1395 |
Instruction Following o3 leads
GPT-4.1 nano: 68.0 (#188), o3: 73.0 (#122)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| IFEval | 84.3% | 86.9% |
| LMArena Instruction Following | 1267 | 1368 |
Long Context o3 leads
GPT-4.1 nano: 23.7 (#291), o3: 53.2 (#6)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| Fiction.LiveBench | 25% | 88.9% |
| LMArena Longer Query | 1283 | 1372 |
| CL-bench | — | 17.8% |
Writing & Preference o3 leads
GPT-4.1 nano: 40.7 (#238), o3: 63.7 (#62)
| Benchmark | GPT-4.1 nano | o3 |
|---|---|---|
| LMArena Text | 1285 | 1410 |
| LMArena Creative Writing | 1260 | 1359 |
| EQ-Bench Creative Writing | 946 | 1676 |
| WildBench | 81.2% | 86.1% |
| LMArena Multi-Turn | 1277 | 1405 |
| Short-Story Creative Writing | — | 83.9% |
Frequently asked questions
Is GPT-4.1 nano better than o3?
o3 is the stronger model overall, scoring 47.4 to 27.7 on the Noometry Index. GPT-4.1 nano costs 20× less per token, which makes it the better buy when o3's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 nano or o3?
GPT-4.1 nano is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; o3 lists at $2 and $8.
Is GPT-4.1 nano or o3 better for coding?
o3 scores higher on coding benchmarks: 47.3 versus 23.8 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 nano does, with 1.05M tokens against 200K.
How many benchmarks do GPT-4.1 nano and o3 share?
37 benchmarks have published results for both models. GPT-4.1 nano has 38 scored results on Noometry and o3 has 63.