Model comparison
Claude Opus 4 vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 43.1 on the Noometry Index.
Last verified . 15 shared benchmarks.
StepFunSummary
- They share 15 benchmarks with published results for both. Claude Opus 4 scores higher in 2 categories and Step 5 Preview in 7 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 5 Preview leads 40.0 to 27.3.
- The biggest single-benchmark swing is CritPt: 0.3% for Claude Opus 4 and 20.9% for Step 5 Preview.
- Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- Step 5 Preview accepts more context: 1.02M tokens versus 200K.
Side by side
| Claude Opus 4 | Step 5 Preview | |
|---|---|---|
| Provider | Anthropic | StepFun |
| Noometry Index | 43.1 | 47.9 |
| Released | 2025-05-22 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | 200K | 1.02M |
| Max output | 32K | 66K |
| Input $ / M tokens | $15 | $1 |
| Output $ / M tokens | $75 | $2.70 |
| Results tracked | 56 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Claude Opus 4: 47.2 (#62), Step 5 Preview: 51.8 (#36)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1442 | 1480 |
| SWE-bench Verified | 70.7% | — |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| GSO | 6.9% | — |
| WeirdML | 43.7% | — |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Not comparable
Claude Opus 4: 34.8 (#42), Step 5 Preview: —
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
Reasoning Step 5 Preview leads
Claude Opus 4: 27.3 (#121), Step 5 Preview: 40.0 (#57)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| CritPt | 0.3% | 20.9% |
| LMArena Hard Prompts | 1399 | 1465 |
| ARC-AGI-2 | 8.6% | — |
| SimpleBench | 58.8% | — |
| Kagi LLM Benchmark | 74.3% | — |
| ARC-AGI-1 | 35.7% | — |
| EnigmaEval | 5.6% | — |
| DTBench | 81.6% | — |
| LMCA | 37.4% | — |
| Epoch Capabilities Index | 142.67 | — |
| ForecastBench | 61.1 | — |
Math Step 5 Preview leads
Claude Opus 4: 42.0 (#86), Step 5 Preview: 46.1 (#72)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Math | 1390 | 1470 |
| OTIS Mock AIME 2024-2025 | 64.4% | — |
| ProofBench | — | 42% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Claude Opus 4 leads
Claude Opus 4: 44.0 (#88), Step 5 Preview: 41.2 (#112)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1386 | 1470 |
| GPQA Diamond | 76.3% | — |
| Humanity's Last Exam | 10.7% | — |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 70.8% | — |
Multimodal Step 5 Preview leads
Claude Opus 4: 31.5 (#106), Step 5 Preview: 41.0 (#41)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1192 | 1267 |
| GeoBench | 49% | — |
| VPCT | 38% | — |
Multilingual Step 5 Preview leads
Claude Opus 4: 48.8 (#138), Step 5 Preview: 53.8 (#54)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1362 | 1432 |
| LMArena Chinese | 1386 | 1519 |
| LMArena Russian | 1392 | 1436 |
| LMArena Spanish | 1389 | 1439 |
| LMArena French | 1372 | — |
| LMArena German | 1391 | — |
| LMArena Japanese | 1331 | — |
| LMArena Korean | 1321 | — |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), Step 5 Preview: 76.0 (#50)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1406 | 1444 |
| IFEval | 91.8% | — |
Long Context Step 5 Preview leads
Claude Opus 4: 39.6 (#172), Step 5 Preview: 44.5 (#64)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1422 | 1455 |
| Fiction.LiveBench | 61.1% | — |
Writing & Preference Step 5 Preview leads
Claude Opus 4: 61.2 (#89), Step 5 Preview: 62.8 (#71)
| Benchmark | Claude Opus 4 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1377 | 1442 |
| LMArena Creative Writing | 1387 | 1410 |
| LMArena Multi-Turn | 1396 | 1447 |
| Short-Story Creative Writing | 83.6% | — |
| EQ-Bench Creative Writing | 1580 | — |
| WildBench | 85.2% | — |
Frequently asked questions
Is Claude Opus 4 better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 43.1 on the Noometry Index.
Which is cheaper, Claude Opus 4 or Step 5 Preview?
Step 5 Preview is cheaper. It lists at $1 per million input tokens and $2.70 per million output tokens; Claude Opus 4 lists at $15 and $75.
Is Claude Opus 4 or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 47.2 in the Noometry coding category.
Which has the bigger context window?
Step 5 Preview does, with 1.02M tokens against 200K.
How many benchmarks do Claude Opus 4 and Step 5 Preview share?
15 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Step 5 Preview has 18.