Qwen3.6 Plus vs Step 3.7 Flash
Updated October 21, 2026. 17 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Only Qwen3.6 Plus has a coding score, 44.1 on the Vibe Coding Index; Step 3.7 Flash has none, so the two cannot be ranked on coding.
Step 3.7 Flash has the edge where output tokens cost 1.7x less, and where what matters is ALE-Bench or τ²-Bench Telecom, where it leads.
- Shared results
- 17
- Qwen3.6 Plus only
- 66
- Step 3.7 Flash only
- 1
- Like-for-like categories
- 6 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Only Qwen3.6 Plus has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Qwen3.6 Plus
Qwen3.6 Plus leads, 27.0 against 19.5.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Qwen3.6 Plus has a GPQA Diamond result, 87.4%.
Directional only
Math
AIME
No call
Only Qwen3.6 Plus has a AIME result, 94.6%.
Directional only
Knowledge
MMLU Pro
No call
Only Qwen3.6 Plus has a MMLU Pro result, 87.7%.
Directional only
Web design
Design Arena: websites
Qwen3.6 Plus
Qwen3.6 Plus leads, 1243 against 1196.
Same board, both measured
Long documents
Listed context window
Qwen3.6 Plus
Qwen3.6 Plus takes 1M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Step 3.7 Flash
Qwen3.6 Plus $0.0013, Step 3.7 Flash $0.0008 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Qwen3.6 Plus has the number, $1.19.
Directional only
At a glance
- Coding index
- Qwen3.6 Plus 44.1 from 3 benchmarks, Step 3.7 Flash not benchmarked.
- Widest category
- Coding, 8.0 points apart on average over 3 shared benchmarks.
- Widest single gap
- AA Coding Index: Qwen3.6 Plus 54.5 against Step 3.7 Flash 39.6, 14.9 points apart on a 0 to 100 scale.
- Output price
- Qwen3.6 Plus $1.95 against Step 3.7 Flash $1.15 per million output tokens, 1.7x apart.
- Context window
- Qwen3.6 Plus takes 1M tokens, Step 3.7 Flash 262K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Qwen3.6 Plus | Bars | Step 3.7 Flash | Ties |
|---|---|---|---|---|
| Coding | 2 | 1 | ||
| Intelligence | 1 | 0 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 2 | 0 | ||
| Web design | 7 | 1 | ||
| General | 2 | 0 | ||
| All shared boards | 14 | 3 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index21
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| LMArena WebDev | 1461 | Not measured | Only Qwen3.6 Plus measured |
| AA Coding Index | 54.5 | 39.6 | Qwen3.6 Plus |
| LiveCodeBench (Vals AI) | 86.0% | Not measured | Only Qwen3.6 Plus measured |
| SWE-bench Verified (Vals AI) | 73.4% | Not measured | Only Qwen3.6 Plus measured |
| Vibe Code Bench (Vals AI) | 25.6% | Not measured | Only Qwen3.6 Plus measured |
| Code migration (Vals AI) | 11.1% | Not measured | Only Qwen3.6 Plus measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only Qwen3.6 Plus measured |
| Terminal-Bench 2.1 (Vals AI) | 53.2% | Not measured | Only Qwen3.6 Plus measured |
| SciCode | Not measured | 43.9% | Only Step 3.7 Flash measured |
| Terminal-Bench Hard (Artificial Analysis run) | 43.9% | 35.6% | Qwen3.6 Plus |
| ALE-Bench | 670.1 | 694.1 | Step 3.7 Flash |
| LMArena Text: coding | 1467 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: software and IT | 1462 | Not measured | Only Qwen3.6 Plus measured |
| LMArena WebDev: React | 1454 | Not measured | Only Qwen3.6 Plus measured |
| LMArena WebDev: HTML | 1463 | Not measured | Only Qwen3.6 Plus measured |
| LMArena WebDev: from an image | 1472 | Not measured | Only Qwen3.6 Plus measured |
| Neither model has a result on the other 22 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| AA Intelligence Index | 27.0 | 19.5 | Qwen3.6 Plus |
| Epoch Capabilities Index | 147.3 | Not measured | Only Qwen3.6 Plus measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| Finance Agent (Vals AI) | 40.9% | Not measured | Only Qwen3.6 Plus measured |
| Tax Agent Bench (Vals AI) | 7.7% | Not measured | Only Qwen3.6 Plus measured |
| Legal Research Bench (Vals AI) | 14.9% | Not measured | Only Qwen3.6 Plus measured |
| Excel Modeling Benchmark (Vals AI) | 32.9% | Not measured | Only Qwen3.6 Plus measured |
| τ²-Bench Telecom | 97.7% | 98.5% | Step 3.7 Flash |
| Vending-Bench 2 | $5,115 | Not measured | Only Qwen3.6 Plus measured |
| Neither model has a result on the other 33 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| GPQA Diamond | 87.4% | Not measured | Only Qwen3.6 Plus measured |
| MMMU Pro (Vals AI) | 84.2% | Not measured | Only Qwen3.6 Plus measured |
| GPQA Diamond (Artificial Analysis run) | 88.2% | 80.9% | Qwen3.6 Plus |
| Humanity's Last Exam (Artificial Analysis run) | 27.8% | 21.4% | Qwen3.6 Plus |
| GPQA Diamond (Epoch AI run) | 88.4% | Not measured | Only Qwen3.6 Plus measured |
| Chess puzzles | 14.0% | Not measured | Only Qwen3.6 Plus measured |
| Mystery game puzzles | 12.0% | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: hard prompts | 1448 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: expert | 1454 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: hard prompts in English | 1455 | Not measured | Only Qwen3.6 Plus measured |
| Neither model has a result on the other 11 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| AIME | 94.6% | Not measured | Only Qwen3.6 Plus measured |
| OTIS Mock AIME | 93.3% | Not measured | Only Qwen3.6 Plus measured |
| FrontierMath, tiers 1 to 3 | 38.2% | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: math | 1449 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: mathematical work | 1448 | Not measured | Only Qwen3.6 Plus measured |
| Neither model has a result on the other 8 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| MMLU Pro | 87.7% | Not measured | Only Qwen3.6 Plus measured |
| LegalBench (Vals AI) | 84.2% | Not measured | Only Qwen3.6 Plus measured |
| CaseLaw (Vals AI) | 51.5% | Not measured | Only Qwen3.6 Plus measured |
| CorpFin (Vals AI) | 61.9% | Not measured | Only Qwen3.6 Plus measured |
| TaxEval (Vals AI) | 74.7% | Not measured | Only Qwen3.6 Plus measured |
| MortgageTax (Vals AI) | 68.0% | Not measured | Only Qwen3.6 Plus measured |
| MedCode (Vals AI) | 36.9% | Not measured | Only Qwen3.6 Plus measured |
| SimpleQA Verified | 44.1% | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: science | 1439 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: medicine and healthcare | 1442 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: legal and government | 1437 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: business and finance | 1430 | Not measured | Only Qwen3.6 Plus measured |
| Neither model has a result on the other 5 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index71
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| Design Arena: websites | 1243 | 1196 | Qwen3.6 Plus |
| Design Arena: UI components | 1233 | 1179 | Qwen3.6 Plus |
| Design Arena: game dev | 1211 | 1158 | Qwen3.6 Plus |
| Design Arena: data viz | 1228 | 1172 | Qwen3.6 Plus |
| Design Arena: 3D | 1202 | 1129 | Qwen3.6 Plus |
| Design Arena: all code categories | 1236 | 1182 | Qwen3.6 Plus |
| Design Arena: SVG | 1161 | 1085 | Qwen3.6 Plus |
| Design Arena: ASCII art | 1114 | 1152 | Step 3.7 Flash |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 44.9% | Not measured | Only Qwen3.6 Plus measured |
| MedScribe (Vals AI) | 77.0% | Not measured | Only Qwen3.6 Plus measured |
| Vals Multimodal Index | 51.5% | Not measured | Only Qwen3.6 Plus measured |
| IFBench | 75.2% | 67.3% | Qwen3.6 Plus |
| Long context reasoning (AA-LCR) | 78.3% | 73.7% | Qwen3.6 Plus |
| DTBench | 81.9% | Not measured | Only Qwen3.6 Plus measured |
| LMCA | 33.1% | Not measured | Only Qwen3.6 Plus measured |
| CL-bench | 20.3% | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text | 1436 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: instruction following | 1424 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: creative writing | 1404 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: English | 1445 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: non-English | 1424 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: Chinese | 1477 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: German | 1451 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: Japanese | 1390 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: Korean | 1379 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: French | 1455 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: Spanish | 1432 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: Russian | 1434 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: Polish | 1433 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: multi-turn | 1438 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: longer queries | 1439 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: writing and language | 1418 | Not measured | Only Qwen3.6 Plus measured |
| LMArena Text: entertainment and media | 1402 | Not measured | Only Qwen3.6 Plus measured |
| Neither model has a result on the other 21 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3.6 Plus | Step 3.7 Flash | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Step 3.7 Flash costs less
Repository review
50K input + 3K output tokens
Step 3.7 Flash costs less
Agent loop
200K input + 10K output tokens
Step 3.7 Flash costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Qwen3.6 Plus | StepFun Step 3.7 Flash |
|---|---|---|
| Maker | Qwen | StepFun |
| Weights | No public weights listed | Open weights |
| Released | April 2, 2026 | May 28, 2026 |
| Context window | 1M | 262K |
| Input $/Mtok | $0.325 | $0.20 |
| Output $/Mtok | $1.95 | $1.15 |
| Coding index | 44.1 | Not benchmarked |
| Independent measurements behind it | 3 | 0 |
| Cost per solved task | $1.19 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3.6 Plus better than Step 3.7 Flash for coding?
Only Qwen3.6 Plus has a coding score, 44.1 on the Vibe Coding Index; Step 3.7 Flash has none, so the two cannot be ranked on coding. Step 3.7 Flash has the edge where output tokens cost 1.7x less, and where what matters is ALE-Bench or τ²-Bench Telecom, where it leads.
Which is cheaper, Qwen3.6 Plus or Step 3.7 Flash?
Qwen3.6 Plus costs $0.325 per million input tokens and $1.95 per million output tokens; Step 3.7 Flash costs $0.20 and $1.15. Cost per solved task is known for Qwen3.6 Plus only.
Which has the bigger context window, Qwen3.6 Plus or Step 3.7 Flash?
Qwen3.6 Plus, with 1M tokens against 262K.
How many benchmarks is the Qwen3.6 Plus vs Step 3.7 Flash comparison based on?
17 benchmarks have a result for both models as of October 21, 2026. Qwen3.6 Plus's coding index rests on 3 independent benchmarks and Step 3.7 Flash's on 0. Where only one model has a result, the page says so instead of naming a winner.