Model comparison
Qwen3 Max vs Step 3.7 Flash
Qwen3 Max is the stronger model overall, scoring 43.7 to 37.3 on the Noometry Index. Step 3.7 Flash costs 5.8× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Alibaba (Qwen)
StepFunSummary
- They share 2 benchmarks with published results for both. Qwen3 Max scores higher in 2 categories and Step 3.7 Flash in 1 category; 2 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 3.7 Flash leads 42.9 to 38.7.
- The biggest single-benchmark swing is NYT Connections (extended): 30.1% for Qwen3 Max 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 $1.20 / $6 for Qwen3 Max.
- Qwen3 Max accepts more context: 262K tokens versus 256K.
- Step 3.7 Flash has downloadable open weights; the other is API-only.
Side by side
| Qwen3 Max | Step 3.7 Flash | |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 43.7 | 37.3 |
| Released | 2025-09-23 | 2026-05-29 |
| Weights | Proprietary | Open |
| Context window | 262K | 256K |
| Max output | 66K | 256K |
| Input $ / M tokens | $1.20 | $0.18 |
| Output $ / M tokens | $6 | $1.11 |
| Results tracked | 33 | 5 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3 Max leads
Qwen3 Max: 43.0 (#93), Step 3.7 Flash: 40.0 (#150)
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| ALE-Bench | 370.45 | 694.12 |
| SciCode | — | 40% |
| LMArena Coding | 1456 | — |
Agentic & Tool Use Not comparable
Qwen3 Max: —, Step 3.7 Flash: —
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| Vending-Bench 2 | 71.56 | — |
Reasoning Too close to call
Qwen3 Max: 22.6 (#190), Step 3.7 Flash: 21.6 (#219)
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| NYT Connections (extended) | 30.1% | 39.7% |
| Kagi LLM Benchmark | 72.5% | — |
| CritPt | — | 2.3% |
| Chess Puzzles | 4% | — |
| LMArena Hard Prompts | 1448 | — |
| Mystery Game Puzzles | 5% | — |
| DTBench | 82.1% | — |
| LMCA | 28.3% | — |
| Epoch Capabilities Index | 142.38 | — |
Math Step 3.7 Flash leads
Qwen3 Max: 38.7 (#131), Step 3.7 Flash: 42.9 (#82)
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| FrontierMath (Tiers 1-3) | 18.9% | — |
| MathArena Final-Answer Competitions | — | 68.5% |
| OTIS Mock AIME 2024-2025 | 73.3% | — |
| LMArena Math | 1446 | — |
| MATH Level 5 | 97.1% | — |
Knowledge Not comparable
Qwen3 Max: 48.1 (#78), Step 3.7 Flash: —
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| GPQA Diamond | 72.6% | — |
| SimpleQA Verified | 48.7% | — |
| LMArena Expert | 1455 | — |
Multilingual Not comparable
Qwen3 Max: 53.7 (#62), Step 3.7 Flash: —
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| LMArena Non-English | 1429 | — |
| LMArena Chinese | 1478 | — |
| LMArena French | 1449 | — |
| LMArena German | 1463 | — |
| LMArena Japanese | 1397 | — |
| LMArena Korean | 1399 | — |
| LMArena Russian | 1428 | — |
| LMArena Spanish | 1462 | — |
Instruction Following Not comparable
Qwen3 Max: 74.8 (#87), Step 3.7 Flash: —
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| LMArena Instruction Following | 1419 | — |
Long Context Not comparable
Qwen3 Max: 41.6 (#134), Step 3.7 Flash: —
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| Fiction.LiveBench | 66.7% | — |
| CL-bench | 14.5% | — |
| LMArena Longer Query | 1438 | — |
Writing & Preference Not comparable
Qwen3 Max: 62.4 (#76), Step 3.7 Flash: —
| Benchmark | Qwen3 Max | Step 3.7 Flash |
|---|---|---|
| LMArena Text | 1439 | — |
| LMArena Creative Writing | 1402 | — |
| LMArena Multi-Turn | 1446 | — |
Frequently asked questions
Is Qwen3 Max better than Step 3.7 Flash?
Qwen3 Max is the stronger model overall, scoring 43.7 to 37.3 on the Noometry Index. Step 3.7 Flash costs 5.8× less per token, which makes it the better buy when Qwen3 Max's lead doesn't matter for your workload.
Which is cheaper, Qwen3 Max 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 Max lists at $1.20 and $6.
Is Qwen3 Max or Step 3.7 Flash better for coding?
Qwen3 Max scores higher on coding benchmarks: 43.0 versus 40.0 in the Noometry coding category.
Which has the bigger context window?
Qwen3 Max does, with 262K tokens against 256K.
How many benchmarks do Qwen3 Max and Step 3.7 Flash share?
2 benchmarks have published results for both models. Qwen3 Max has 33 scored results on Noometry and Step 3.7 Flash has 5.