Model comparison
GPT-4.1 nano vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 27.9 on the Noometry Index. GPT-4.1 nano costs 8.1× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
StepFunSummary
- They share 15 benchmarks with published results for both. GPT-4.1 nano scores higher in 0 categories and Step 5 Preview in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 5 Preview leads 40.0 to 8.5.
- The biggest single-benchmark swing is SciCode: 25.9% for GPT-4.1 nano and 58.9% for Step 5 Preview.
- GPT-4.1 nano is cheaper at $0.10 / $0.40 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- GPT-4.1 nano accepts more context: 1.05M tokens versus 1.02M.
Side by side
| GPT-4.1 nano | Step 5 Preview | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 27.9 | 47.9 |
| Released | 2025-04-14 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 1.02M |
| Max output | 33K | 66K |
| Input $ / M tokens | $0.10 | $1 |
| Output $ / M tokens | $0.40 | $2.70 |
| Results tracked | 38 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
GPT-4.1 nano: 24.1 (#330), Step 5 Preview: 51.8 (#36)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| SciCode | 25.9% | 58.9% |
| LMArena Coding | 1306 | 1480 |
| Aider Polyglot | 8.9% | — |
| LMArena WebDev | — | 1564 |
| WeirdML | 19% | — |
Agentic & Tool Use Not comparable
GPT-4.1 nano: 26.5 (#104), Step 5 Preview: —
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 33% | — |
Reasoning Step 5 Preview leads
GPT-4.1 nano: 8.5 (#349), Step 5 Preview: 40.0 (#57)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| CritPt | 0% | 20.9% |
| LMArena Hard Prompts | 1286 | 1465 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 33.3% | — |
| ARC-AGI-1 | 0% | — |
| DTBench | 52.5% | — |
| LMCA | 5.5% | — |
| Epoch Capabilities Index | 129.62 | — |
Math Step 5 Preview leads
GPT-4.1 nano: 26.9 (#252), Step 5 Preview: 46.1 (#72)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| LMArena Math | 1274 | 1470 |
| OTIS Mock AIME 2024-2025 | 28.9% | — |
| ProofBench | — | 42% |
| Omni-MATH | 36.7% | — |
| MATH Level 5 | 70% | — |
| FrontierMath (Feb 2025 set) | 1% | — |
Knowledge Step 5 Preview leads
GPT-4.1 nano: 21.8 (#273), Step 5 Preview: 41.2 (#112)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1272 | 1470 |
| GPQA Diamond | 48.9% | — |
| SimpleQA Verified | 6% | — |
| MMLU-Pro | 55% | — |
| GPQA (HELM) | 50.7% | — |
Multimodal Step 5 Preview leads
GPT-4.1 nano: 29.2 (#113), Step 5 Preview: 41.0 (#41)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1063 | 1267 |
Multilingual Step 5 Preview leads
GPT-4.1 nano: 41.6 (#205), Step 5 Preview: 53.8 (#54)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1260 | 1432 |
| LMArena Chinese | 1270 | 1519 |
| LMArena Russian | 1261 | 1436 |
| LMArena German | 1288 | — |
| LMArena Japanese | 1198 | — |
| LMArena Spanish | — | 1439 |
Instruction Following Step 5 Preview leads
GPT-4.1 nano: 67.8 (#193), Step 5 Preview: 76.0 (#50)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1267 | 1444 |
| IFEval | 84.3% | — |
Long Context Step 5 Preview leads
GPT-4.1 nano: 23.7 (#296), Step 5 Preview: 44.5 (#64)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1283 | 1455 |
| Fiction.LiveBench | 25% | — |
Writing & Preference Step 5 Preview leads
GPT-4.1 nano: 40.5 (#243), Step 5 Preview: 62.8 (#71)
| Benchmark | GPT-4.1 nano | Step 5 Preview |
|---|---|---|
| LMArena Text | 1285 | 1442 |
| LMArena Creative Writing | 1260 | 1410 |
| LMArena Multi-Turn | 1277 | 1447 |
| EQ-Bench Creative Writing | 946 | — |
| WildBench | 81.2% | — |
Frequently asked questions
Is GPT-4.1 nano better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 27.9 on the Noometry Index. GPT-4.1 nano costs 8.1× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Which is cheaper, GPT-4.1 nano or Step 5 Preview?
GPT-4.1 nano is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is GPT-4.1 nano or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 24.1 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 nano does, with 1.05M tokens against 1.02M.
How many benchmarks do GPT-4.1 nano and Step 5 Preview share?
15 benchmarks have published results for both models. GPT-4.1 nano has 38 scored results on Noometry and Step 5 Preview has 18.