Qwen3 235B A22B Thinking 2507 vs Qwen3 30B A3B
Qwen3 235B A22B Thinking 2507 leads Qwen3 30B A3B on the Artificial Analysis Intelligence Index, 12.7 to 7.6, and Qwen3 30B A3B costs 7.0x less per output token ($0.50 vs $3.50 per million). Updated October 22, 2026; 40 benchmarks have a result for both.
Qwen3 235B A22B Thinking 2507 or Qwen3 30B A3B: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Qwen3 235B A22B Thinking 2507 for a larger context window (128K against 41K tokens) and work measured by Design Arena: UI components or Design Arena: data viz, where it leads.
- Shared results
- 40
- Qwen3 235B A22B Thinking 2507 only
- 6
- Qwen3 30B A3B only
- 13
- Like-for-like categories
- 7 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
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
Qwen3 235B A22B Thinking 2507
Qwen3 235B A22B Thinking 2507 leads, 12.7 against 7.6.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Qwen3 235B A22B Thinking 2507 has a AA Agentic Index result, 1.3.
Directional only
Web design
Design Arena: websites
Qwen3 235B A22B Thinking 2507
Qwen3 235B A22B Thinking 2507 leads, 1052 against 954.
Same board, both measured
Long documents
Listed context window
Qwen3 235B A22B Thinking 2507
Qwen3 235B A22B Thinking 2507 takes 128K tokens against 41K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Qwen3 30B A3B
Qwen3 235B A22B Thinking 2507 $0.0022, Qwen3 30B A3B $0.0004 per turn.
List prices
At a glance
- Coding index
- Qwen3 235B A22B Thinking 2507 not benchmarked, Qwen3 30B A3B not benchmarked.
- Widest category
- Reasoning, 19.3 points apart on average over 4 shared benchmarks.
- Widest single gap
- Fiction.LiveBench: Qwen3 235B A22B Thinking 2507 75.0% against Qwen3 30B A3B 40.6%, 34.4 points apart on a 0 to 100 scale.
- Output price
- Qwen3 235B A22B Thinking 2507 $3.50 against Qwen3 30B A3B $0.50 per million output tokens, 7.0x apart.
- Context window
- Qwen3 235B A22B Thinking 2507 takes 128K tokens, Qwen3 30B A3B 41K.
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 30B A3B | Ties |
|---|---|---|---|---|
| Coding | 3 | 0 | ||
| Intelligence | 1 | 0 | ||
| Reasoning | 4 | 0 | ||
| Math | 3 | 0 | ||
| Knowledge | 4 | 0 | ||
| Web design | 3 | 2 | ||
| General | 20 | 0 | ||
| All shared boards | 38 | 2 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| AA Coding Index | 22.1 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 50.6% | Only Qwen3 30B A3B measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 6.8% | Only Qwen3 30B A3B measured |
| WeirdML | 41.0% | 29.8% | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: coding | 1424 | 1337 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: software and IT | 1431 | 1340 | Qwen3 235B A22B Thinking 2507 |
| Neither model has a result on the other 37 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| AA Intelligence Index | 12.7 | 7.6 | Qwen3 235B A22B Thinking 2507 |
| Epoch Capabilities Index | Not measured | 137.3 | Only Qwen3 30B A3B measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| AA Agentic Index | 1.3 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| τ²-Bench Telecom | Not measured | 26.0% | Only Qwen3 30B A3B measured |
| Vending-Bench 2 | -$11 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Neither model has a result on the other 39 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 61.6% | Only Qwen3 30B A3B measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 6.2% | Only Qwen3 30B A3B measured |
| GPQA Diamond (Epoch AI run) | 80.1% | 61.7% | Qwen3 235B A22B Thinking 2507 |
| Mystery game puzzles | 9.0% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: hard prompts | 1416 | 1314 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: expert | 1463 | 1314 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: hard prompts in English | 1432 | 1329 | Qwen3 235B A22B Thinking 2507 |
| Neither model has a result on the other 16 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| AA Math Index | Not measured | 72.3 | Only Qwen3 30B A3B measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 72.3% | Only Qwen3 30B A3B measured |
| OTIS Mock AIME | 86.7% | 62.8% | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: math | 1412 | 1355 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: mathematical work | 1435 | 1363 | Qwen3 235B A22B Thinking 2507 |
| MathArena | Not measured | 9.4% | Only Qwen3 30B A3B measured |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | Not measured | 77.7% | Only Qwen3 30B A3B measured |
| SimpleQA Verified | 40.4% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: science | 1433 | 1321 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: medicine and healthcare | 1456 | 1318 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: legal and government | 1415 | 1314 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: business and finance | 1413 | 1309 | Qwen3 235B A22B Thinking 2507 |
| Neither model has a result on the other 14 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index32
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| Design Arena: websites | 1052 | 954 | Qwen3 235B A22B Thinking 2507 |
| Design Arena: UI components | 942 | 943 | Qwen3 30B A3B |
| Design Arena: game dev | 956 | 898 | Qwen3 235B A22B Thinking 2507 |
| Design Arena: data viz | 950 | 967 | Qwen3 30B A3B |
| Design Arena: 3D | 997 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Design Arena: all code categories | 1032 | 941 | Qwen3 235B A22B Thinking 2507 |
| Neither model has a result on the other 3 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index200
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | Leader |
|---|---|---|---|
| IFBench | Not measured | 41.5% | Only Qwen3 30B A3B measured |
| Long context reasoning (AA-LCR) | Not measured | 0.0% | Only Qwen3 30B A3B measured |
| DTBench | 80.3% | 60.3% | Qwen3 235B A22B Thinking 2507 |
| LMCA | 29.3% | 15.8% | Qwen3 235B A22B Thinking 2507 |
| Fiction.LiveBench | 75.0% | 40.6% | Qwen3 235B A22B Thinking 2507 |
| Creative writing (Lech Mazur) | 82.4% | 75.3% | Qwen3 235B A22B Thinking 2507 |
| LMArena Text | 1415 | 1317 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: instruction following | 1386 | 1283 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: creative writing | 1384 | 1271 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: English | 1429 | 1335 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: non-English | 1397 | 1295 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: Chinese | 1481 | 1347 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: German | 1408 | 1307 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: Japanese | 1386 | 1254 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: Korean | 1375 | 1261 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: French | Not measured | 1352 | Only Qwen3 30B A3B measured |
| LMArena Text: Spanish | 1389 | 1317 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: Russian | 1401 | 1291 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: Polish | 1390 | 1299 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: multi-turn | 1406 | 1307 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: longer queries | 1400 | 1312 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: writing and language | 1388 | 1290 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: entertainment and media | 1370 | 1258 | Qwen3 235B A22B Thinking 2507 |
| 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
| Benchmark | Qwen3 235B A22B Thinking 2507 | Qwen3 30B A3B | 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 30B A3B costs less
Repository review
50K input + 3K output tokens
Qwen3 30B A3B costs less
Agent loop
200K input + 10K output tokens
Qwen3 30B A3B 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 30B A3B |
|---|---|---|
| Maker | Qwen | Qwen |
| Weights | Open weights | Open weights |
| Released | July 25, 2025 | April 28, 2025 |
| Context window | 128K | 41K |
| Input $/Mtok | $0.45 | $0.12 |
| Output $/Mtok | $3.50 | $0.50 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3 235B A22B Thinking 2507 better than Qwen3 30B A3B for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Qwen3 235B A22B Thinking 2507 for a larger context window (128K against 41K tokens) and work measured by Design Arena: UI components or Design Arena: data viz, where it leads.
Which is cheaper, Qwen3 235B A22B Thinking 2507 or Qwen3 30B A3B?
Qwen3 235B A22B Thinking 2507 costs $0.45 per million input tokens and $3.50 per million output tokens; Qwen3 30B A3B costs $0.12 and $0.50. Cost per solved task is known for neither of them.
Which has the bigger context window, Qwen3 235B A22B Thinking 2507 or Qwen3 30B A3B?
Qwen3 235B A22B Thinking 2507, with 128K tokens against 41K.
How many benchmarks is the Qwen3 235B A22B Thinking 2507 vs Qwen3 30B A3B comparison based on?
40 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 30B A3B's on 0. Where only one model has a result, the page says so instead of naming a winner.