Qwen3 235B A22B Thinking 2507 vs Qwen3.7 Max
Qwen3.7 Max leads Qwen3 235B A22B Thinking 2507 on the Artificial Analysis Intelligence Index, 29.5 to 12.7, and Qwen3 235B A22B Thinking 2507 costs 1.3x less per output token ($3.50 vs $4.425 per million). Updated October 22, 2026; 15 benchmarks have a result for both.
Qwen3 235B A22B Thinking 2507 or Qwen3.7 Max: which to pick
Only Qwen3.7 Max has a coding score, 54.1 on the Vibe Coding Index; Qwen3 235B A22B Thinking 2507 has none, so the two cannot be ranked on coding.
Pick Qwen3 235B A22B Thinking 2507 for output tokens 1.3x cheaper ($3.50 vs $4.425 per million).
- Shared results
- 15
- Qwen3 235B A22B Thinking 2507 only
- 31
- Qwen3.7 Max only
- 45
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
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 Qwen3.7 Max has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Qwen3.7 Max
Qwen3.7 Max leads, 29.5 against 12.7.
Same board, both measured
Agentic work
AA Agentic Index
Qwen3.7 Max
Qwen3.7 Max leads, 22.5 against 1.3.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Qwen3.7 Max has a GPQA Diamond result, 90.2%.
Directional only
Knowledge
MMLU Pro
No call
Only Qwen3.7 Max has a MMLU Pro result, 89.3%.
Directional only
Web design
Design Arena: websites
Qwen3.7 Max
Qwen3.7 Max leads, 1274 against 1052.
Same board, both measured
Long documents
Listed context window
Qwen3.7 Max
Qwen3.7 Max takes 1M tokens against 128K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 235B A22B Thinking 2507
Qwen3 235B A22B Thinking 2507 $0.0022, Qwen3.7 Max $0.0037 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Qwen3.7 Max has the number, $4.69.
Directional only
At a glance
- Coding index
- Qwen3 235B A22B Thinking 2507 not benchmarked, Qwen3.7 Max 54.1 from 3 benchmarks.
- Widest category
- Coding, 43.9 points apart on average over 1 shared benchmark.
- Widest single gap
- AA Coding Index: Qwen3 235B A22B Thinking 2507 22.1 against Qwen3.7 Max 66.0, 43.9 points apart on a 0 to 100 scale.
- Output price
- Qwen3 235B A22B Thinking 2507 $3.50 against Qwen3.7 Max $4.425 per million output tokens, 1.3x apart.
- Context window
- Qwen3 235B A22B Thinking 2507 takes 128K tokens, Qwen3.7 Max 1M.
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 235B A22B Thinking 2507 | Bars | Qwen3.7 Max | Ties |
|---|---|---|---|---|
| Coding | 0 | 1 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 0 | 2 | ||
| Math | 0 | 1 | ||
| Knowledge | 0 | 1 | ||
| Web design | 0 | 6 | ||
| General | 0 | 2 | ||
| All shared boards | 0 | 15 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index01
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| LMArena WebDev | Not measured | 1515 | Only Qwen3.7 Max measured |
| AA Coding Index | 22.1 | 66.0 | Qwen3.7 Max |
| LiveCodeBench (Vals AI) | Not measured | 87.1% | Only Qwen3.7 Max measured |
| SWE-bench Verified (Vals AI) | Not measured | 68.8% | Only Qwen3.7 Max measured |
| Vibe Code Bench (Vals AI) | Not measured | 47.7% | Only Qwen3.7 Max measured |
| Code migration (Vals AI) | Not measured | 13.1% | Only Qwen3.7 Max measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 61.0% | Only Qwen3.7 Max measured |
| SciCode | Not measured | 49.5% | Only Qwen3.7 Max measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 50.8% | Only Qwen3.7 Max measured |
| WeirdML | 41.0% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| ALE-Bench | Not measured | 1189.4 | Only Qwen3.7 Max measured |
| LMArena Text: coding | 1424 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: software and IT | 1431 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena WebDev: React | Not measured | 1514 | Only Qwen3.7 Max measured |
| LMArena WebDev: HTML | Not measured | 1512 | Only Qwen3.7 Max measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| AA Intelligence Index | 12.7 | 29.5 | Qwen3.7 Max |
| Epoch Capabilities Index | Not measured | 154.0 | Only Qwen3.7 Max measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| AA Agentic Index | 1.3 | 22.5 | Qwen3.7 Max |
| Design Arena: agents building web apps | Not measured | 1208 | Only Qwen3.7 Max measured |
| Design Arena: agents building full-stack apps | Not measured | 1191 | Only Qwen3.7 Max measured |
| Design Arena: agents building mobile apps | Not measured | 1154 | Only Qwen3.7 Max measured |
| Design Arena: agents building Android apps | Not measured | 1159 | Only Qwen3.7 Max measured |
| Design Arena: agents building Godot games | Not measured | 1187 | Only Qwen3.7 Max measured |
| Design Arena: agents building games | Not measured | 1161 | Only Qwen3.7 Max measured |
| Design Arena: HTML slides | Not measured | 1098 | Only Qwen3.7 Max measured |
| Design Arena: PowerPoint slides | Not measured | 1169 | Only Qwen3.7 Max measured |
| Finance Agent (Vals AI) | Not measured | 47.8% | Only Qwen3.7 Max measured |
| Tax Agent Bench (Vals AI) | Not measured | 15.8% | Only Qwen3.7 Max measured |
| Legal Research Bench (Vals AI) | Not measured | 25.5% | Only Qwen3.7 Max measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 57.0% | Only Qwen3.7 Max measured |
| τ²-Bench Telecom | Not measured | 94.7% | Only Qwen3.7 Max measured |
| Vending-Bench 2 | -$11 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 90.2% | Only Qwen3.7 Max measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 92.3% | Only Qwen3.7 Max measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 40.5% | Only Qwen3.7 Max measured |
| GPQA Diamond (Epoch AI run) | 80.1% | 90.9% | Qwen3.7 Max |
| SimpleBench | Not measured | 70.4% | Only Qwen3.7 Max measured |
| Chess puzzles | Not measured | 26.0% | Only Qwen3.7 Max measured |
| Mystery game puzzles | 9.0% | 32.0% | Qwen3.7 Max |
| EBR-bench | Not measured | 9.5% | Only Qwen3.7 Max measured |
| LMArena Text: hard prompts | 1416 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: expert | 1463 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: hard prompts in English | 1432 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Neither model has a result on the other 12 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| OTIS Mock AIME | 86.7% | 95.6% | Qwen3.7 Max |
| FrontierMath, tiers 1 to 3 | Not measured | 64.6% | Only Qwen3.7 Max measured |
| FrontierMath, tier 4 | Not measured | 34.1% | Only Qwen3.7 Max measured |
| ProofBench | Not measured | 26.0% | Only Qwen3.7 Max measured |
| LMArena Text: math | 1412 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: mathematical work | 1435 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 89.3% | Only Qwen3.7 Max measured |
| LegalBench (Vals AI) | Not measured | 84.9% | Only Qwen3.7 Max measured |
| CorpFin (Vals AI) | Not measured | 63.7% | Only Qwen3.7 Max measured |
| TaxEval (Vals AI) | Not measured | 75.3% | Only Qwen3.7 Max measured |
| MedCode (Vals AI) | Not measured | 38.8% | Only Qwen3.7 Max measured |
| SimpleQA Verified | 40.4% | 55.8% | Qwen3.7 Max |
| LMArena Text: science | 1433 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: medicine and healthcare | 1456 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: legal and government | 1415 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: business and finance | 1413 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Neither model has a result on the other 10 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| Design Arena: websites | 1052 | 1274 | Qwen3.7 Max |
| Design Arena: UI components | 942 | 1267 | Qwen3.7 Max |
| Design Arena: game dev | 956 | 1265 | Qwen3.7 Max |
| Design Arena: data viz | 950 | 1276 | Qwen3.7 Max |
| Design Arena: 3D | 997 | 1265 | Qwen3.7 Max |
| Design Arena: all code categories | 1032 | 1275 | Qwen3.7 Max |
| Design Arena: SVG | Not measured | 1216 | Only Qwen3.7 Max measured |
| Design Arena: ASCII art | Not measured | 1211 | Only Qwen3.7 Max measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | Leader |
|---|---|---|---|
| MedScribe (Vals AI) | Not measured | 79.4% | Only Qwen3.7 Max measured |
| IFBench | Not measured | 80.5% | Only Qwen3.7 Max measured |
| Long context reasoning (AA-LCR) | Not measured | 79.0% | Only Qwen3.7 Max measured |
| DTBench | 80.3% | 92.3% | Qwen3.7 Max |
| LMCA | 29.3% | 44.0% | Qwen3.7 Max |
| Fiction.LiveBench | 75.0% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Creative writing (Lech Mazur) | 82.4% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text | 1415 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: instruction following | 1386 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: creative writing | 1384 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: English | 1429 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: non-English | 1397 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: Chinese | 1481 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: German | 1408 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: Japanese | 1386 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: Korean | 1375 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: Spanish | 1389 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: Russian | 1401 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: Polish | 1390 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: multi-turn | 1406 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: longer queries | 1400 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: writing and language | 1388 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: entertainment and media | 1370 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| EQ-Bench 4 | Not measured | 1110 | Only Qwen3.7 Max measured |
| Neither model has a result on the other 24 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3.7 Max | 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
Qwen3 235B A22B Thinking 2507 costs less
Repository review
50K input + 3K output tokens
Qwen3 235B A22B Thinking 2507 costs less
Agent loop
200K input + 10K output tokens
Qwen3 235B A22B Thinking 2507 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 235B A22B Thinking 2507 | Qwen3.7 Max |
|---|---|---|
| Maker | Qwen | Qwen |
| Weights | Open weights | No public weights listed |
| Released | July 25, 2025 | May 21, 2026 |
| Context window | 128K | 1M |
| Input $/Mtok | $0.45 | $1.475 |
| Output $/Mtok | $3.50 | $4.425 |
| Coding index | Not benchmarked | 54.1 |
| Independent measurements behind it | 0 | 3 |
| Cost per solved task | Not computable | $4.69 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3 235B A22B Thinking 2507 better than Qwen3.7 Max for coding?
Only Qwen3.7 Max has a coding score, 54.1 on the Vibe Coding Index; Qwen3 235B A22B Thinking 2507 has none, so the two cannot be ranked on coding. Pick Qwen3 235B A22B Thinking 2507 for output tokens 1.3x cheaper ($3.50 vs $4.425 per million).
Which is cheaper, Qwen3 235B A22B Thinking 2507 or Qwen3.7 Max?
Qwen3 235B A22B Thinking 2507 costs $0.45 per million input tokens and $3.50 per million output tokens; Qwen3.7 Max costs $1.475 and $4.425. Cost per solved task is known for Qwen3.7 Max only.
Which has the bigger context window, Qwen3 235B A22B Thinking 2507 or Qwen3.7 Max?
Qwen3.7 Max, with 1M tokens against 128K.
How many benchmarks is the Qwen3 235B A22B Thinking 2507 vs Qwen3.7 Max comparison based on?
15 benchmarks have a result for both models as of October 22, 2026. Qwen3 235B A22B Thinking 2507's coding index rests on 0 independent benchmarks and Qwen3.7 Max's on 3. Where only one model has a result, the page says so instead of naming a winner.