Model comparison
Claude Sonnet 5 vs Step 5 Preview
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 47.9 on the Noometry Index. Step 5 Preview costs 2.8× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Last verified . 18 shared benchmarks.
StepFunSummary
- They share 18 benchmarks with published results for both. Claude Sonnet 5 scores higher in 8 categories and Step 5 Preview in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 5 leads 66.2 to 46.1.
- The biggest single-benchmark swing is ProofBench: 77% for Claude Sonnet 5 and 42% for Step 5 Preview.
- Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Step 5 Preview accepts more context: 1.02M tokens versus 1M.
Side by side
| Claude Sonnet 5 | Step 5 Preview | |
|---|---|---|
| Provider | Anthropic | StepFun |
| Noometry Index | 54.6 | 47.9 |
| Released | 2026-06-29 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.02M |
| Max output | 128K | 66K |
| Input $ / M tokens | $2 | $1 |
| Output $ / M tokens | $10 | $2.70 |
| Results tracked | 51 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5 leads
Claude Sonnet 5: 55.5 (#26), Step 5 Preview: 51.8 (#36)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| LMArena WebDev | 1541 | 1564 |
| SciCode | 54.3% | 58.9% |
| LMArena Coding | 1483 | 1480 |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| CursorBench | 34.1% | — |
| GSO | 37.3% | — |
| WeirdML | 68.8% | — |
| ALE-Bench | 1,463 | — |
Agentic & Tool Use Not comparable
Claude Sonnet 5: 42.8 (#18), Step 5 Preview: —
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| APEX-Agents | 54.5% | — |
| GBAEval | 65.3% | — |
| LMArena Search | 1194 | — |
| Vending-Bench 2 | 6,378 | — |
Reasoning Claude Sonnet 5 leads
Claude Sonnet 5: 49.1 (#39), Step 5 Preview: 40.0 (#57)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| CritPt | 16.9% | 20.9% |
| LMArena Hard Prompts | 1461 | 1465 |
| SimpleBench | 60.6% | — |
| NYT Connections (extended) | 75.1% | — |
| Chess Puzzles | 35% | — |
| Mystery Game Puzzles | 35% | — |
| DTBench | 92.5% | — |
| LMCA | 50% | — |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| Epoch Capabilities Index | 156.21 | — |
| ForecastBench | 61.1 | — |
Math Claude Sonnet 5 leads
Claude Sonnet 5: 66.2 (#27), Step 5 Preview: 46.1 (#72)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| ProofBench | 77% | 42% |
| LMArena Math | 1467 | 1470 |
| FrontierMath (Tiers 1-3) | 65.6% | — |
| FrontierMath Tier 4 | 29.3% | — |
| OTIS Mock AIME 2024-2025 | 94.7% | — |
Knowledge Claude Sonnet 5 leads
Claude Sonnet 5: 55.6 (#47), Step 5 Preview: 41.2 (#112)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1490 | 1470 |
| GPQA Diamond | 90.5% | — |
| SimpleQA Verified | 33.7% | — |
Multimodal Claude Sonnet 5 leads
Claude Sonnet 5: 42.4 (#31), Step 5 Preview: 41.0 (#41)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1274 | 1267 |
| Blueprint-Bench 2 | 24.9% | — |
| LMArena Document | 1466 | — |
Multilingual Too close to call
Claude Sonnet 5: 53.8 (#55), Step 5 Preview: 53.8 (#54)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1431 | 1432 |
| LMArena Chinese | 1477 | 1519 |
| LMArena Russian | 1451 | 1436 |
| LMArena Spanish | 1437 | 1439 |
| LMArena French | 1460 | — |
| LMArena German | 1440 | — |
| LMArena Japanese | 1422 | — |
| LMArena Korean | 1411 | — |
Instruction Following Too close to call
Claude Sonnet 5: 76.3 (#41), Step 5 Preview: 76.0 (#50)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1452 | 1444 |
Long Context Too close to call
Claude Sonnet 5: 44.8 (#55), Step 5 Preview: 44.5 (#64)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1463 | 1455 |
Writing & Preference Claude Sonnet 5 leads
Claude Sonnet 5: 69.2 (#25), Step 5 Preview: 62.8 (#71)
| Benchmark | Claude Sonnet 5 | Step 5 Preview |
|---|---|---|
| LMArena Text | 1442 | 1442 |
| LMArena Creative Writing | 1416 | 1410 |
| LMArena Multi-Turn | 1454 | 1447 |
| EQ-Bench Creative Writing | 1794 | — |
| EQ-Bench 4 | 1236 | — |
Frequently asked questions
Is Claude Sonnet 5 better than Step 5 Preview?
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 47.9 on the Noometry Index. Step 5 Preview costs 2.8× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Which is cheaper, Claude Sonnet 5 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 Sonnet 5 lists at $2 and $10.
Is Claude Sonnet 5 or Step 5 Preview better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 51.8 in the Noometry coding category.
Which has the bigger context window?
Step 5 Preview does, with 1.02M tokens against 1M.
How many benchmarks do Claude Sonnet 5 and Step 5 Preview share?
18 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and Step 5 Preview has 18.