Model comparison
Qwen3.7 Plus vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 45.3 on the Noometry Index. Qwen3.7 Plus costs 2.0× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 16 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 16 benchmarks with published results for both. Qwen3.7 Plus scores higher in 5 categories and Step 5 Preview in 4 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 5 Preview leads 51.8 to 36.6.
- The biggest single-benchmark swing is SciCode: 45.5% for Qwen3.7 Plus and 58.9% for Step 5 Preview.
- Qwen3.7 Plus is cheaper at $0.40 / $1.60 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Step 5 Preview accepts more context: 1.02M tokens versus 1M.
Side by side
| Qwen3.7 Plus | Step 5 Preview | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 45.3 | 47.9 |
| Released | 2026-06-02 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.02M |
| Max output | 131K | 66K |
| Input $ / M tokens | $0.40 | $1 |
| Output $ / M tokens | $1.60 | $2.70 |
| Results tracked | 32 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Qwen3.7 Plus: 36.6 (#206), Step 5 Preview: 51.8 (#36)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| SciCode | 45.5% | 58.9% |
| LMArena Coding | 1473 | 1480 |
| FrontierCode | 10.2% | — |
| LMArena WebDev | — | 1564 |
Agentic & Tool Use Not comparable
Qwen3.7 Plus: 21.4 (#138), Step 5 Preview: —
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| OSWorld 2.0 | 2.8% | — |
Reasoning Too close to call
Qwen3.7 Plus: 39.3 (#59), Step 5 Preview: 40.0 (#57)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| CritPt | 9.1% | 20.9% |
| LMArena Hard Prompts | 1460 | 1465 |
| NYT Connections (extended) | 74.8% | — |
| Chess Puzzles | 24% | — |
| Mystery Game Puzzles | 17% | — |
| DTBench | 84% | — |
| LMCA | 37.6% | — |
| Epoch Capabilities Index | 147.37 | — |
Math Qwen3.7 Plus leads
Qwen3.7 Plus: 50.5 (#56), Step 5 Preview: 46.1 (#72)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| LMArena Math | 1466 | 1470 |
| FrontierMath (Tiers 1-3) | 34.4% | — |
| OTIS Mock AIME 2024-2025 | 93.3% | — |
| ProofBench | — | 42% |
Knowledge Qwen3.7 Plus leads
Qwen3.7 Plus: 54.9 (#51), Step 5 Preview: 41.2 (#112)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1467 | 1470 |
| GPQA Diamond | 87.9% | — |
Multimodal Too close to call
Qwen3.7 Plus: 41.8 (#33), Step 5 Preview: 41.0 (#41)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1279 | 1267 |
| LMArena Document | 1444 | — |
Multilingual Too close to call
Qwen3.7 Plus: 54.8 (#38), Step 5 Preview: 53.8 (#54)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1445 | 1432 |
| LMArena Chinese | 1510 | 1519 |
| LMArena Russian | 1457 | 1436 |
| LMArena Spanish | 1457 | 1439 |
| LMArena French | 1473 | — |
| LMArena German | 1471 | — |
| LMArena Japanese | 1413 | — |
| LMArena Korean | 1415 | — |
Instruction Following Too close to call
Qwen3.7 Plus: 75.8 (#52), Step 5 Preview: 76.0 (#50)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1440 | 1444 |
Long Context Too close to call
Qwen3.7 Plus: 44.5 (#65), Step 5 Preview: 44.5 (#64)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1455 | 1455 |
Writing & Preference Qwen3.7 Plus leads
Qwen3.7 Plus: 64.3 (#56), Step 5 Preview: 62.8 (#71)
| Benchmark | Qwen3.7 Plus | Step 5 Preview |
|---|---|---|
| LMArena Text | 1455 | 1442 |
| LMArena Creative Writing | 1439 | 1410 |
| LMArena Multi-Turn | 1460 | 1447 |
Frequently asked questions
Is Qwen3.7 Plus better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 45.3 on the Noometry Index. Qwen3.7 Plus costs 2.0× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Which is cheaper, Qwen3.7 Plus or Step 5 Preview?
Qwen3.7 Plus is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is Qwen3.7 Plus or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 36.6 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 Qwen3.7 Plus and Step 5 Preview share?
16 benchmarks have published results for both models. Qwen3.7 Plus has 32 scored results on Noometry and Step 5 Preview has 18.