Model comparison
Qwen3.6 Plus vs Step 3.7 Flash
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 37.3 on the Noometry Index. Step 3.7 Flash costs 2.7× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.
Last verified . 4 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 4 benchmarks with published results for both. Qwen3.6 Plus scores higher in 3 categories and Step 3.7 Flash in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.6 Plus leads 51.8 to 42.9.
- The biggest single-benchmark swing is NYT Connections (extended): 60.3% for Qwen3.6 Plus and 39.7% for Step 3.7 Flash.
- Step 3.7 Flash is cheaper at $0.18 / $1.11 per million input/output tokens, against $0.50 / $3 for Qwen3.6 Plus.
- Qwen3.6 Plus accepts more context: 1M tokens versus 256K.
- Step 3.7 Flash has downloadable open weights; the other is API-only.
Side by side
| Qwen3.6 Plus | Step 3.7 Flash | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 47.5 | 37.3 |
| Released | 2026-03-31 | 2026-05-29 |
| Weights | Proprietary | Open |
| Context window | 1M | 256K |
| Max output | 66K | 256K |
| Input $ / M tokens | $0.50 | $0.18 |
| Output $ / M tokens | $3 | $1.11 |
| Results tracked | 37 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Qwen3.6 Plus: 40.8 (#130), Step 3.7 Flash: 40.0 (#150)
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| SciCode | 40.7% | 40% |
| ALE-Bench | 670.15 | 694.12 |
| SWE-bench Verified | 57.9% | — |
| LMArena WebDev | 1461 | — |
| LMArena Coding | 1467 | — |
Agentic & Tool Use Not comparable
Qwen3.6 Plus: —, Step 3.7 Flash: —
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| Vending-Bench 2 | 5,115 | — |
Reasoning Qwen3.6 Plus leads
Qwen3.6 Plus: 29.3 (#93), Step 3.7 Flash: 21.6 (#219)
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| NYT Connections (extended) | 60.3% | 39.7% |
| CritPt | 2.9% | 2.3% |
| Chess Puzzles | 17% | — |
| Thematic Generalization | 59.5% | — |
| LMArena Hard Prompts | 1449 | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 81.9% | — |
| LMCA | 33.1% | — |
| Epoch Capabilities Index | 147.65 | — |
Math Qwen3.6 Plus leads
Qwen3.6 Plus: 51.8 (#54), Step 3.7 Flash: 42.9 (#82)
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 38.2% | — |
| MathArena Final-Answer Competitions | — | 68.5% |
| OTIS Mock AIME 2024-2025 | 93.3% | — |
| LMArena Math | 1450 | — |
| FrontierMath (Feb 2025 set) | 26.2% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |
Knowledge Not comparable
Qwen3.6 Plus: 56.1 (#45), Step 3.7 Flash: —
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| GPQA Diamond | 88.4% | — |
| SimpleQA Verified | 44.1% | — |
| LMArena Expert | 1454 | — |
Multilingual Not comparable
Qwen3.6 Plus: 53.3 (#70), Step 3.7 Flash: —
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1424 | — |
| LMArena Chinese | 1477 | — |
| LMArena French | 1455 | — |
| LMArena German | 1452 | — |
| LMArena Japanese | 1389 | — |
| LMArena Korean | 1379 | — |
| LMArena Russian | 1434 | — |
| LMArena Spanish | 1432 | — |
Instruction Following Not comparable
Qwen3.6 Plus: 75.0 (#74), Step 3.7 Flash: —
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1425 | — |
Long Context Not comparable
Qwen3.6 Plus: 45.2 (#49), Step 3.7 Flash: —
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| CL-bench | 20.3% | — |
| LMArena Longer Query | 1439 | — |
Writing & Preference Not comparable
Qwen3.6 Plus: 62.2 (#82), Step 3.7 Flash: —
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash |
|---|---|---|
| LMArena Text | 1437 | — |
| LMArena Creative Writing | 1404 | — |
| LMArena Multi-Turn | 1438 | — |
Frequently asked questions
Is Qwen3.6 Plus better than Step 3.7 Flash?
Qwen3.6 Plus is the stronger model overall, scoring 47.5 to 37.3 on the Noometry Index. Step 3.7 Flash costs 2.7× less per token, which makes it the better buy when Qwen3.6 Plus's lead doesn't matter for your workload.
Which is cheaper, Qwen3.6 Plus or Step 3.7 Flash?
Step 3.7 Flash is cheaper. It lists at $0.18 per million input tokens and $1.11 per million output tokens; Qwen3.6 Plus lists at $0.50 and $3.
Is Qwen3.6 Plus or Step 3.7 Flash better for coding?
They score almost the same on coding (40.8 vs 40.0); test both on your own repository before choosing.
Which has the bigger context window?
Qwen3.6 Plus does, with 1M tokens against 256K.
How many benchmarks do Qwen3.6 Plus and Step 3.7 Flash share?
4 benchmarks have published results for both models. Qwen3.6 Plus has 37 scored results on Noometry and Step 3.7 Flash has 5.