Enter opens the highlighted result. Esc closes.

Qwen3 235B A22B Thinking 2507 vs Qwen3.6 Plus

Qwen3.6 Plus leads Qwen3 235B A22B Thinking 2507 on the Artificial Analysis Intelligence Index, 27.0 to 12.7, and Qwen3.6 Plus costs 1.8x less per output token ($1.95 vs $3.50 per million). Updated October 22, 2026; 42 benchmarks have a result for both.

Replace Qwen3 235B A22B Thinking 2507 Replace Qwen3.6 Plus
Qwen3 235B A22B Thinking 2507Qwen
Coding
No data
Overall
52.9
Output $/M
$3.50
3
Qwen3.6 PlusQwen
Coding
44.1
Overall
56.2
Output $/M
$1.95
39
vs boards won of 42 shared

Qwen3 235B A22B Thinking 2507 or Qwen3.6 Plus: which to pick

Only Qwen3.6 Plus has a coding score, 44.1 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 work measured by LMArena Text: expert or LMArena Text: medicine and healthcare, where it leads.

Model AQwen3 235B A22B Thinking 2507QwenNot benchmarkedNo coding benchmarkNo cost per solved taskModel BQwen3.6 PlusQwen44.1/1003 independent benchmarks$1.19 per solved task
Shared results
42
Qwen3 235B A22B Thinking 2507 only
4
Qwen3.6 Plus only
41
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, Qwen3.6 Plus 44.1 from 3 benchmarks.
Widest category
Agentic work, 32.0 points apart on average over 1 shared benchmark.
Widest single gap
Design Arena: UI components: Qwen3 235B A22B Thinking 2507 942 against Qwen3.6 Plus 1230, 36.0 points apart on a 0 to 100 scale.
Output price
Qwen3 235B A22B Thinking 2507 $3.50 against Qwen3.6 Plus $1.95 per million output tokens, 1.8x apart.
Context window
Qwen3 235B A22B Thinking 2507 takes 128K tokens, Qwen3.6 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 2507BarsQwen3.6 PlusTies
Coding03
Intelligence01
Agentic work01
Reasoning14
Math03
Knowledge14
Web design06
General117
All shared boards339
CodingLike-for-likeSix of these boards feed the Vibe Coding Index03
Coding benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
LMArena WebDevNot measured1461Only Qwen3.6 Plus measured
AA Coding Index22.154.5Qwen3.6 Plus
LiveCodeBench (Vals AI)Not measured86.0%Only Qwen3.6 Plus measured
SWE-bench Verified (Vals AI)Not measured73.4%Only Qwen3.6 Plus measured
Vibe Code Bench (Vals AI)Not measured25.6%Only Qwen3.6 Plus measured
Code migration (Vals AI)Not measured11.1%Only Qwen3.6 Plus measured
ProgramBench (Vals AI)Not measured0.0%Only Qwen3.6 Plus measured
Terminal-Bench 2.1 (Vals AI)Not measured53.2%Only Qwen3.6 Plus measured
Terminal-Bench Hard (Artificial Analysis run)Not measured43.9%Only Qwen3.6 Plus measured
WeirdML41.0%Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
ALE-BenchNot measured670.1Only Qwen3.6 Plus measured
LMArena Text: coding14241467Qwen3.6 Plus
LMArena Text: software and IT14311462Qwen3.6 Plus
LMArena WebDev: ReactNot measured1454Only Qwen3.6 Plus measured
LMArena WebDev: HTMLNot measured1463Only Qwen3.6 Plus measured
LMArena WebDev: from an imageNot measured1472Only Qwen3.6 Plus measured
Neither model has a result on the other 27 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
AA Intelligence Index12.727.0Qwen3.6 Plus
Epoch Capabilities IndexNot measured147.3Only Qwen3.6 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 Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
AA Agentic Index1.3Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Finance Agent (Vals AI)Not measured40.9%Only Qwen3.6 Plus measured
Tax Agent Bench (Vals AI)Not measured7.7%Only Qwen3.6 Plus measured
Legal Research Bench (Vals AI)Not measured14.9%Only Qwen3.6 Plus measured
Excel Modeling Benchmark (Vals AI)Not measured32.9%Only Qwen3.6 Plus measured
τ²-Bench TelecomNot measured97.7%Only Qwen3.6 Plus measured
Vending-Bench 2-$11$5,115Qwen3.6 Plus
Neither model has a result on the other 35 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index14
Reasoning benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
GPQA DiamondNot measured87.4%Only Qwen3.6 Plus measured
MMMU Pro (Vals AI)Not measured84.2%Only Qwen3.6 Plus measured
GPQA Diamond (Artificial Analysis run)Not measured88.2%Only Qwen3.6 Plus measured
Humanity's Last Exam (Artificial Analysis run)Not measured27.8%Only Qwen3.6 Plus measured
GPQA Diamond (Epoch AI run)80.1%88.4%Qwen3.6 Plus
Chess puzzlesNot measured14.0%Only Qwen3.6 Plus measured
Mystery game puzzles9.0%12.0%Qwen3.6 Plus
LMArena Text: hard prompts14161448Qwen3.6 Plus
LMArena Text: expert14631454Qwen3 235B A22B Thinking 2507
LMArena Text: hard prompts in English14321455Qwen3.6 Plus
Neither model has a result on the other 13 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index03
Math benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
AIMENot measured94.6%Only Qwen3.6 Plus measured
OTIS Mock AIME86.7%93.3%Qwen3.6 Plus
FrontierMath, tiers 1 to 3Not measured38.2%Only Qwen3.6 Plus measured
LMArena Text: math14121449Qwen3.6 Plus
LMArena Text: mathematical work14351448Qwen3.6 Plus
Neither model has a result on the other 9 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index14
Knowledge benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
MMLU ProNot measured87.7%Only Qwen3.6 Plus measured
LegalBench (Vals AI)Not measured84.2%Only Qwen3.6 Plus measured
CaseLaw (Vals AI)Not measured51.5%Only Qwen3.6 Plus measured
CorpFin (Vals AI)Not measured61.9%Only Qwen3.6 Plus measured
TaxEval (Vals AI)Not measured74.7%Only Qwen3.6 Plus measured
MortgageTax (Vals AI)Not measured68.0%Only Qwen3.6 Plus measured
MedCode (Vals AI)Not measured36.9%Only Qwen3.6 Plus measured
SimpleQA Verified40.4%44.1%Qwen3.6 Plus
LMArena Text: science14331439Qwen3.6 Plus
LMArena Text: medicine and healthcare14561442Qwen3 235B A22B Thinking 2507
LMArena Text: legal and government14151437Qwen3.6 Plus
LMArena Text: business and finance14131430Qwen3.6 Plus
Neither model has a result on the other 8 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 Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
Design Arena: websites10521243Qwen3.6 Plus
Design Arena: UI components9421230Qwen3.6 Plus
Design Arena: game dev9561211Qwen3.6 Plus
Design Arena: data viz9501228Qwen3.6 Plus
Design Arena: 3D9971202Qwen3.6 Plus
Design Arena: all code categories10321235Qwen3.6 Plus
Design Arena: SVGNot measured1160Only Qwen3.6 Plus measured
Design Arena: ASCII artNot measured1113Only Qwen3.6 Plus measured
Neither model has a result on the other 1 board here
GeneralLike-for-likePublic boards, shown as published, kept out of the index117
General benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 PlusLeader
SAGE (Vals AI)Not measured44.9%Only Qwen3.6 Plus measured
MedScribe (Vals AI)Not measured77.0%Only Qwen3.6 Plus measured
Vals Multimodal IndexNot measured51.5%Only Qwen3.6 Plus measured
IFBenchNot measured75.2%Only Qwen3.6 Plus measured
Long context reasoning (AA-LCR)Not measured78.3%Only Qwen3.6 Plus measured
DTBench80.3%81.9%Qwen3.6 Plus
LMCA29.3%33.1%Qwen3.6 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
CL-benchNot measured20.3%Only Qwen3.6 Plus measured
LMArena Text14151436Qwen3.6 Plus
LMArena Text: instruction following13861424Qwen3.6 Plus
LMArena Text: creative writing13841404Qwen3.6 Plus
LMArena Text: English14291445Qwen3.6 Plus
LMArena Text: non-English13971424Qwen3.6 Plus
LMArena Text: Chinese14811477Qwen3 235B A22B Thinking 2507
LMArena Text: German14081451Qwen3.6 Plus
LMArena Text: Japanese13861390Qwen3.6 Plus
LMArena Text: Korean13751379Qwen3.6 Plus
LMArena Text: FrenchNot measured1455Only Qwen3.6 Plus measured
LMArena Text: Spanish13891432Qwen3.6 Plus
LMArena Text: Russian14011434Qwen3.6 Plus
LMArena Text: Polish13901433Qwen3.6 Plus
LMArena Text: multi-turn14061438Qwen3.6 Plus
LMArena Text: longer queries14001439Qwen3.6 Plus
LMArena Text: writing and language13881418Qwen3.6 Plus
LMArena Text: entertainment and media13701402Qwen3.6 Plus
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
Image generation benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.6 Plus
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.6 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
Qwen3.6 Plus$0.0013Fits in one request

Qwen3.6 Plus costs less

Repository review

50K input + 3K output tokens

Qwen3 235B A22B Thinking 2507$0.033Fits in one request
Qwen3.6 Plus$0.022Fits in one request

Qwen3.6 Plus costs less

Agent loop

200K input + 10K output tokens

Qwen3 235B A22B Thinking 2507$0.125Exceeds the context window
Qwen3.6 Plus$0.085Fits in one request

Qwen3.6 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 Qwen3.6 Plus
SpecificationQwen3 235B A22B Thinking 2507Qwen3.6 Plus
MakerQwenQwen
WeightsOpen weightsNo public weights listed
ReleasedJuly 25, 2025April 2, 2026
Context window128K1M
Input $/Mtok$0.45$0.325
Output $/Mtok$3.50$1.95
Coding indexNot benchmarked44.1
Independent measurements behind it03
Cost per solved taskNot computable$1.19
SourceOpenRouter listingOpenRouter listing

Questions

Is Qwen3 235B A22B Thinking 2507 better than Qwen3.6 Plus for coding?

Only Qwen3.6 Plus has a coding score, 44.1 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 work measured by LMArena Text: expert or LMArena Text: medicine and healthcare, where it leads.

Which is cheaper, Qwen3 235B A22B Thinking 2507 or Qwen3.6 Plus?

Qwen3 235B A22B Thinking 2507 costs $0.45 per million input tokens and $3.50 per million output tokens; Qwen3.6 Plus costs $0.325 and $1.95. Cost per solved task is known for Qwen3.6 Plus only.

Which has the bigger context window, Qwen3 235B A22B Thinking 2507 or Qwen3.6 Plus?

Qwen3.6 Plus, with 1M tokens against 128K.

How many benchmarks is the Qwen3 235B A22B Thinking 2507 vs Qwen3.6 Plus comparison based on?

42 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 Qwen3.6 Plus's on 3. Where only one model has a result, the page says so instead of naming a winner.

Compare next