Enter opens the highlighted result. Esc closes.

Qwen3 235B A22B Thinking 2507 vs Qwen3.7 Plus

Qwen3.7 Plus leads Qwen3 235B A22B Thinking 2507 on the Artificial Analysis Intelligence Index, 25.2 to 12.7, and Qwen3.7 Plus costs 2.7x less per output token ($1.28 vs $3.50 per million). Updated October 22, 2026; 41 benchmarks have a result for both.

Replace Qwen3 235B A22B Thinking 2507 Replace Qwen3.7 Plus
Qwen3 235B A22B Thinking 2507Qwen
Coding
No data
Overall
52.9
Output $/M
$3.50
0
Qwen3.7 PlusQwen
Coding
33.2
Overall
54.0
Output $/M
$1.28
41
vs boards won of 41 shared

Qwen3 235B A22B Thinking 2507 or Qwen3.7 Plus: which to pick

Only Qwen3.7 Plus has a coding score, 33.2 on the Vibe Coding Index; Qwen3 235B A22B Thinking 2507 has none, so the two cannot be ranked on coding.

Compare them on price, context and the boards below.

Model AQwen3 235B A22B Thinking 2507QwenNot benchmarkedNo coding benchmarkNo cost per solved taskModel BQwen3.7 PlusQwen33.2/1002 independent benchmarksNo cost per solved task
Shared results
41
Qwen3 235B A22B Thinking 2507 only
5
Qwen3.7 Plus only
34
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.

At a glance

Coding index
Qwen3 235B A22B Thinking 2507 not benchmarked, Qwen3.7 Plus 33.2 from 2 benchmarks.
Widest category
Web design, 34.2 points apart on average over 6 shared benchmarks.
Widest single gap
Design Arena: data viz: Qwen3 235B A22B Thinking 2507 950 against Qwen3.7 Plus 1277, 40.9 points apart on a 0 to 100 scale.
Output price
Qwen3 235B A22B Thinking 2507 $3.50 against Qwen3.7 Plus $1.28 per million output tokens, 2.7x apart.
Context window
Qwen3 235B A22B Thinking 2507 takes 128K tokens, Qwen3.7 Plus 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.

Boards won, where both have a result
CategoryQwen3 235B A22B Thinking 2507BarsQwen3.7 PlusTies
Coding03
Intelligence01
Agentic work01
Reasoning05
Math03
Knowledge04
Web design06
General018
All shared boards041
CodingLike-for-likeSix of these boards feed the Vibe Coding Index03
Coding benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
AA Coding Index22.155.9Qwen3.7 Plus
Vibe Code Bench (Vals AI)Not measured46.4%Only Qwen3.7 Plus measured
Code migration (Vals AI)Not measured12.9%Only Qwen3.7 Plus measured
Terminal-Bench 2.1 (Vals AI)Not measured52.8%Only Qwen3.7 Plus measured
SkillsBench (Vals AI)Not measured54.3%Only Qwen3.7 Plus measured
SciCodeNot measured46.1%Only Qwen3.7 Plus measured
Terminal-Bench Hard (Artificial Analysis run)Not measured47.0%Only Qwen3.7 Plus measured
FrontierCodeNot measured10.2%Only Qwen3.7 Plus measured
WeirdML41.0%Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: coding14241473Qwen3.7 Plus
LMArena Text: software and IT14311473Qwen3.7 Plus
Neither model has a result on the other 32 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
AA Intelligence Index12.725.2Qwen3.7 Plus
Epoch Capabilities IndexNot measured147.0Only Qwen3.7 Plus measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
AA Agentic Index1.317.5Qwen3.7 Plus
Finance Agent (Vals AI)Not measured38.2%Only Qwen3.7 Plus measured
Tax Agent Bench (Vals AI)Not measured10.5%Only Qwen3.7 Plus measured
Legal Research Bench (Vals AI)Not measured16.4%Only Qwen3.7 Plus measured
Excel Modeling Benchmark (Vals AI)Not measured49.3%Only Qwen3.7 Plus measured
τ²-Bench TelecomNot measured93.0%Only Qwen3.7 Plus measured
OSWorld 2.0Not measured2.8%Only Qwen3.7 Plus measured
Vending-Bench 2-$11Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Neither model has a result on the other 34 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
Reasoning benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
GPQA Diamond (Artificial Analysis run)Not measured90.0%Only Qwen3.7 Plus measured
Humanity's Last Exam (Artificial Analysis run)Not measured35.6%Only Qwen3.7 Plus measured
GPQA Diamond (Epoch AI run)80.1%87.9%Qwen3.7 Plus
Mystery game puzzles9.0%17.0%Qwen3.7 Plus
LMArena Text: hard prompts14161460Qwen3.7 Plus
LMArena Text: expert14631466Qwen3.7 Plus
LMArena Text: hard prompts in English14321460Qwen3.7 Plus
LMArena Vision: diagramsNot measured1292Only Qwen3.7 Plus measured
LMArena Vision: homeworkNot measured1295Only Qwen3.7 Plus measured
Neither model has a result on the other 14 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index03
Math benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
OTIS Mock AIME86.7%93.3%Qwen3.7 Plus
FrontierMath, tiers 1 to 3Not measured34.4%Only Qwen3.7 Plus measured
LMArena Text: math14121467Qwen3.7 Plus
LMArena Text: mathematical work14351470Qwen3.7 Plus
Neither model has a result on the other 10 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
Knowledge benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
MortgageTax (Vals AI)Not measured66.2%Only Qwen3.7 Plus measured
SimpleQA Verified40.4%Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: science14331464Qwen3.7 Plus
LMArena Text: medicine and healthcare14561464Qwen3.7 Plus
LMArena Text: legal and government14151451Qwen3.7 Plus
LMArena Text: business and finance14131435Qwen3.7 Plus
Neither model has a result on the other 14 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index06
Web design benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
Design Arena: websites10521269Qwen3.7 Plus
Design Arena: UI components9421258Qwen3.7 Plus
Design Arena: game dev9561262Qwen3.7 Plus
Design Arena: data viz9501277Qwen3.7 Plus
Design Arena: 3D9971238Qwen3.7 Plus
Design Arena: all code categories10321268Qwen3.7 Plus
Design Arena: SVGNot measured1207Only Qwen3.7 Plus measured
Design Arena: ASCII artNot measured1143Only Qwen3.7 Plus measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index018
General benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
SAGE (Vals AI)Not measured39.3%Only Qwen3.7 Plus measured
Vals Multimodal IndexNot measured53.9%Only Qwen3.7 Plus measured
IFBenchNot measured78.0%Only Qwen3.7 Plus measured
Long context reasoning (AA-LCR)Not measured73.0%Only Qwen3.7 Plus measured
DTBench80.3%84.0%Qwen3.7 Plus
LMCA29.3%37.6%Qwen3.7 Plus
Fiction.LiveBench75.0%Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Creative writing (Lech Mazur)82.4%Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text14151454Qwen3.7 Plus
LMArena Text: instruction following13861439Qwen3.7 Plus
LMArena Text: creative writing13841438Qwen3.7 Plus
LMArena VisionNot measured1278Only Qwen3.7 Plus measured
LMArena Text: English14291458Qwen3.7 Plus
LMArena Text: non-English13971445Qwen3.7 Plus
LMArena Text: Chinese14811509Qwen3.7 Plus
LMArena Text: German14081471Qwen3.7 Plus
LMArena Text: Japanese13861412Qwen3.7 Plus
LMArena Text: Korean13751413Qwen3.7 Plus
LMArena Text: FrenchNot measured1472Only Qwen3.7 Plus measured
LMArena Text: Spanish13891457Qwen3.7 Plus
LMArena Text: Russian14011457Qwen3.7 Plus
LMArena Text: Polish13901449Qwen3.7 Plus
LMArena Text: multi-turn14061459Qwen3.7 Plus
LMArena Text: longer queries14001454Qwen3.7 Plus
LMArena Text: writing and language13881437Qwen3.7 Plus
LMArena Text: entertainment and media13701422Qwen3.7 Plus
LMArena Vision: EnglishNot measured1280Only Qwen3.7 Plus measured
LMArena Vision: ChineseNot measured1332Only Qwen3.7 Plus measured
LMArena Vision: reading textNot measured1288Only Qwen3.7 Plus measured
LMArena Vision: humorNot measured1261Only Qwen3.7 Plus measured
LMArena Vision: creative writingNot measured1258Only Qwen3.7 Plus measured
LMArena DocumentNot measured1444Only Qwen3.7 Plus measured
Neither model has a result on the other 16 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.7 PlusLeader
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$0.0022Fits in one request
Qwen3.7 Plus$0.0010Fits in one request

Qwen3.7 Plus costs less

Repository review

50K input + 3K output tokens

Qwen3 235B A22B Thinking 2507$0.033Fits in one request
Qwen3.7 Plus$0.020Fits in one request

Qwen3.7 Plus costs less

Agent loop

200K input + 10K output tokens

Qwen3 235B A22B Thinking 2507$0.125Exceeds the context window
Qwen3.7 Plus$0.077Fits in one request

Qwen3.7 Plus 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.

Specifications of Qwen3 235B A22B Thinking 2507 and Qwen3.7 Plus
SpecificationQwen3 235B A22B Thinking 2507Qwen3.7 Plus
MakerQwenQwen
WeightsOpen weightsNo public weights listed
ReleasedJuly 25, 2025June 3, 2026
Context window128K1M
Input $/Mtok$0.45$0.32
Output $/Mtok$3.50$1.28
Coding indexNot benchmarked33.2
Independent measurements behind it02
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Qwen3 235B A22B Thinking 2507 better than Qwen3.7 Plus for coding?

Only Qwen3.7 Plus has a coding score, 33.2 on the Vibe Coding Index; Qwen3 235B A22B Thinking 2507 has none, so the two cannot be ranked on coding. Compare them on price, context and the boards below.

Which is cheaper, Qwen3 235B A22B Thinking 2507 or Qwen3.7 Plus?

Qwen3 235B A22B Thinking 2507 costs $0.45 per million input tokens and $3.50 per million output tokens; Qwen3.7 Plus costs $0.32 and $1.28. Cost per solved task is known for neither of them.

Which has the bigger context window, Qwen3 235B A22B Thinking 2507 or Qwen3.7 Plus?

Qwen3.7 Plus, with 1M tokens against 128K.

How many benchmarks is the Qwen3 235B A22B Thinking 2507 vs Qwen3.7 Plus comparison based on?

41 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 Plus's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next