Model comparison
GPT-4.5 vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.2 on the Noometry Index.
Last verified . 13 shared benchmarks.
StepFunSummary
- They share 13 benchmarks with published results for both. GPT-4.5 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 13.9.
Side by side
| GPT-4.5 | Step 5 Preview | |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 37.2 | 47.9 |
| Released | 2025-02-27 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.02M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $2.70 |
| Results tracked | 42 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
GPT-4.5: 42.2 (#109), Step 5 Preview: 51.8 (#36)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1396 | 1480 |
| Aider Polyglot | 44.9% | — |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| WeirdML | 39.4% | — |
| LiveBench Coding | 75.2% | — |
Agentic & Tool Use Not comparable
GPT-4.5: 27.9 (#97), Step 5 Preview: —
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| Cybench | 17.5% | — |
Reasoning Step 5 Preview leads
GPT-4.5: 13.9 (#330), Step 5 Preview: 40.0 (#57)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1403 | 1465 |
| ARC-AGI-2 | 0.8% | — |
| SimpleBench | 34.5% | — |
| ARC-AGI-1 | 10.3% | — |
| CritPt | — | 20.9% |
| EnigmaEval | 3.2% | — |
| LiveBench Reasoning | 71.1% | — |
| LiveBench Data Analysis | 64.3% | — |
| Epoch Capabilities Index | 136.74 | — |
| ForecastBench | 61.7 | — |
| LiveBench | 69% | — |
Math Step 5 Preview leads
GPT-4.5: 32.6 (#211), Step 5 Preview: 46.1 (#72)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Math | 1412 | 1470 |
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| ProofBench | — | 42% |
| LiveBench Math | 69.3% | — |
| MATH Level 5 | 78.6% | — |
Knowledge Step 5 Preview leads
GPT-4.5: 32.5 (#211), Step 5 Preview: 41.2 (#112)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1394 | 1470 |
| GPQA Diamond | 68.7% | — |
| Humanity's Last Exam | 5.4% | — |
| Confabulations | 13.6% | — |
Multimodal Step 5 Preview leads
GPT-4.5: 37.6 (#71), Step 5 Preview: 41.0 (#41)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1195 | 1267 |
| VPCT | 45% | — |
Multilingual Step 5 Preview leads
GPT-4.5: 52.5 (#83), Step 5 Preview: 53.8 (#54)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1413 | 1432 |
| LMArena Chinese | 1421 | 1519 |
| LMArena Russian | 1419 | 1436 |
| LMArena French | 1418 | — |
| LMArena German | 1457 | — |
| LMArena Japanese | 1416 | — |
| LMArena Korean | 1392 | — |
| LMArena Spanish | — | 1439 |
Instruction Following Step 5 Preview leads
GPT-4.5: 72.6 (#134), Step 5 Preview: 76.0 (#50)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1404 | 1444 |
| LiveBench Instruction Following | 72.3% | — |
Long Context Step 5 Preview leads
GPT-4.5: 40.4 (#155), Step 5 Preview: 44.5 (#64)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1406 | 1455 |
| Fiction.LiveBench | 63.9% | — |
Writing & Preference Step 5 Preview leads
GPT-4.5: 56.9 (#134), Step 5 Preview: 62.8 (#71)
| Benchmark | GPT-4.5 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1417 | 1442 |
| LMArena Creative Writing | 1394 | 1410 |
| LMArena Multi-Turn | 1444 | 1447 |
| Short-Story Creative Writing | 75.6% | — |
| EQ-Bench Creative Writing | 1258 | — |
| LiveBench Language | 61.5% | — |
Frequently asked questions
Is GPT-4.5 better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.2 on the Noometry Index.
Is GPT-4.5 or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 42.2 in the Noometry coding category.
How many benchmarks do GPT-4.5 and Step 5 Preview share?
13 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Step 5 Preview has 18.