Enter opens the highlighted result. Esc closes.

Qwen3 235B A22B Thinking 2507 vs Qwen3.5 Plus 2026-04-20

Qwen3.5 Plus 2026-04-20 costs 1.9x less per output token than Qwen3 235B A22B Thinking 2507 ($1.80 vs $3.50 per million). Updated October 23, 2026; 7 benchmarks have a result for both.

Replace Qwen3 235B A22B Thinking 2507 Replace Qwen3.5 Plus 2026-04-20
Qwen3 235B A22B Thinking 2507Qwen
Coding
No data
Overall
52.9
Output $/M
$3.50
1
Qwen3.5 Plus 2026-04-20Qwen
Coding
39.5
Overall
No data
Output $/M
$1.80
5
vs boards won of 7 shared, 1 tied

Qwen3 235B A22B Thinking 2507 or Qwen3.5 Plus 2026-04-20: which to pick

Only Qwen3.5 Plus 2026-04-20 has a coding score, 39.5 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 SimpleQA Verified, where it leads.

Model AQwen3 235B A22B Thinking 2507QwenNot benchmarkedNo coding benchmarkNo cost per solved taskModel BQwen3.5 Plus 2026-04-20Qwen39.5/1002 independent benchmarks$1.35 per solved task
Shared results
7
Qwen3 235B A22B Thinking 2507 only
39
Qwen3.5 Plus 2026-04-20 only
18
Like-for-like categories
5 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, Qwen3.5 Plus 2026-04-20 39.5 from 2 benchmarks.
Widest category
Knowledge, 15.1 points apart on average over 1 shared benchmark.
Widest single gap
SimpleQA Verified: Qwen3 235B A22B Thinking 2507 40.4% against Qwen3.5 Plus 2026-04-20 25.4%, 15.1 points apart on a 0 to 100 scale.
Output price
Qwen3 235B A22B Thinking 2507 $3.50 against Qwen3.5 Plus 2026-04-20 $1.80 per million output tokens, 1.9x apart.
Context window
Qwen3 235B A22B Thinking 2507 takes 128K tokens, Qwen3.5 Plus 2026-04-20 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 2507BarsQwen3.5 Plus 2026-04-20Ties
Agentic work01
Reasoning02
Math001 tie
Knowledge10
General02
All shared boards151 tie
CodingDirectional onlySix of these boards feed the Vibe Coding Index8 boards, none shared
Coding benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
AA Coding Index22.1Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LiveCodeBench (Vals AI)Not measured85.3%Only Qwen3.5 Plus 2026-04-20 measured
SWE-bench Verified (Vals AI)Not measured71.2%Only Qwen3.5 Plus 2026-04-20 measured
Vibe Code Bench (Vals AI)Not measured15.7%Only Qwen3.5 Plus 2026-04-20 measured
WeirdML41.0%Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
ALE-BenchNot measured608.9Only Qwen3.5 Plus 2026-04-20 measured
LMArena Text: coding1424Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: software and IT1431Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Neither model has a result on the other 35 boards here
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
Intelligence benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
AA Intelligence Index12.7Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Epoch Capabilities IndexNot measured146.2Only Qwen3.5 Plus 2026-04-20 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
AA Agentic Index1.3Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Vending-Bench 2-$11$1Qwen3.5 Plus 2026-04-20
Neither model has a result on the other 40 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
Reasoning benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
GPQA DiamondNot measured87.4%Only Qwen3.5 Plus 2026-04-20 measured
MMMU Pro (Vals AI)Not measured22.8%Only Qwen3.5 Plus 2026-04-20 measured
GPQA Diamond (Epoch AI run)80.1%84.8%Qwen3.5 Plus 2026-04-20
Chess puzzlesNot measured8.0%Only Qwen3.5 Plus 2026-04-20 measured
Mystery game puzzles9.0%17.0%Qwen3.5 Plus 2026-04-20
LMArena Text: hard prompts1416Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: expert1463Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: hard prompts in English1432Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Neither model has a result on the other 15 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index00
Math benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
AIMENot measured86.0%Only Qwen3.5 Plus 2026-04-20 measured
OTIS Mock AIME86.7%86.7%Tie
LMArena Text: math1412Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: mathematical work1435Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Neither model has a result on the other 10 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index10
Knowledge benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
MMLU ProNot measured87.2%Only Qwen3.5 Plus 2026-04-20 measured
LegalBench (Vals AI)Not measured85.1%Only Qwen3.5 Plus 2026-04-20 measured
CaseLaw (Vals AI)Not measured59.7%Only Qwen3.5 Plus 2026-04-20 measured
CorpFin (Vals AI)Not measured65.3%Only Qwen3.5 Plus 2026-04-20 measured
MortgageTax (Vals AI)Not measured60.8%Only Qwen3.5 Plus 2026-04-20 measured
MedQA (Vals AI)Not measured95.2%Only Qwen3.5 Plus 2026-04-20 measured
SimpleQA Verified40.4%25.4%Qwen3 235B A22B Thinking 2507
LMArena Text: science1433Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: medicine and healthcare1456Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: legal and government1415Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: business and finance1412Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Neither model has a result on the other 9 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Web design benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
Design Arena: websites1052Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Design Arena: UI components942Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Design Arena: game dev956Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Design Arena: data viz950Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Design Arena: 3D998Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Design Arena: all code categories1032Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
General benchmarks for Qwen3 235B A22B Thinking 2507 and Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
SAGE (Vals AI)Not measured30.4%Only Qwen3.5 Plus 2026-04-20 measured
DTBench80.3%80.5%Qwen3.5 Plus 2026-04-20
LMCA29.3%36.4%Qwen3.5 Plus 2026-04-20
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 measured19.8%Only Qwen3.5 Plus 2026-04-20 measured
CL-bench LifeNot measured12.4%Only Qwen3.5 Plus 2026-04-20 measured
LMArena Text1415Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: instruction following1386Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: creative writing1384Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: English1429Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: non-English1397Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: Chinese1481Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: German1408Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: Japanese1386Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: Korean1375Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: Spanish1389Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: Russian1400Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: Polish1390Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: multi-turn1406Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: longer queries1399Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: writing and language1388Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
LMArena Text: entertainment and media1370Not measuredOnly Qwen3 235B A22B Thinking 2507 measured
Neither model has a result on the other 25 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 Qwen3.5 Plus 2026-04-20
BenchmarkQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20Leader
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
Qwen3.5 Plus 2026-04-20$0.0012Fits in one request

Qwen3.5 Plus 2026-04-20 costs less

Repository review

50K input + 3K output tokens

Qwen3 235B A22B Thinking 2507$0.033Fits in one request
Qwen3.5 Plus 2026-04-20$0.020Fits in one request

Qwen3.5 Plus 2026-04-20 costs less

Agent loop

200K input + 10K output tokens

Qwen3 235B A22B Thinking 2507$0.125Exceeds the context window
Qwen3.5 Plus 2026-04-20$0.078Fits in one request

Qwen3.5 Plus 2026-04-20 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 Qwen3.5 Plus 2026-04-20
SpecificationQwen3 235B A22B Thinking 2507Qwen3.5 Plus 2026-04-20
MakerQwenQwen
WeightsOpen weightsNo public weights listed
ReleasedJuly 25, 2025April 27, 2026
Context window128K1M
Input $/Mtok$0.45$0.30
Output $/Mtok$3.50$1.80
Coding indexNot benchmarked39.5
Independent measurements behind it02
Cost per solved taskNot computable$1.35
SourceOpenRouter listingOpenRouter listing

Questions

Is Qwen3 235B A22B Thinking 2507 better than Qwen3.5 Plus 2026-04-20 for coding?

Only Qwen3.5 Plus 2026-04-20 has a coding score, 39.5 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 SimpleQA Verified, where it leads.

Which is cheaper, Qwen3 235B A22B Thinking 2507 or Qwen3.5 Plus 2026-04-20?

Qwen3 235B A22B Thinking 2507 costs $0.45 per million input tokens and $3.50 per million output tokens; Qwen3.5 Plus 2026-04-20 costs $0.30 and $1.80. Cost per solved task is known for Qwen3.5 Plus 2026-04-20 only.

Which has the bigger context window, Qwen3 235B A22B Thinking 2507 or Qwen3.5 Plus 2026-04-20?

Qwen3.5 Plus 2026-04-20, with 1M tokens against 128K.

How many benchmarks is the Qwen3 235B A22B Thinking 2507 vs Qwen3.5 Plus 2026-04-20 comparison based on?

7 benchmarks have a result for both models as of October 23, 2026. Qwen3 235B A22B Thinking 2507's coding index rests on 0 independent benchmarks and Qwen3.5 Plus 2026-04-20's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next