Qwen3 235B A22B Thinking 2507 vs Inkling Small
Inkling Small leads Qwen3 235B A22B Thinking 2507 on the Artificial Analysis Intelligence Index, 25.7 to 12.7, and Inkling Small costs 2.9x less per output token ($1.20 vs $3.50 per million). Updated October 22, 2026; 34 benchmarks have a result for both.
Qwen3 235B A22B Thinking 2507 or Inkling Small: which to pick
Only Inkling Small has a coding score, 27.3 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 Mystery game puzzles or LMArena Text: expert, where it leads.
- Shared results
- 34
- Qwen3 235B A22B Thinking 2507 only
- 12
- Inkling Small only
- 50
- 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
Only Inkling Small has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Inkling Small
Inkling Small leads, 25.7 against 12.7.
Same board, both measured
Agentic work
AA Agentic Index
Inkling Small
Inkling Small leads, 23.5 against 1.3.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Inkling Small has a GPQA Diamond result, 83.6%.
Directional only
Knowledge
MMLU Pro
No call
Only Inkling Small has a MMLU Pro result, 85.6%.
Directional only
Web design
Design Arena: websites
No call
Only Qwen3 235B A22B Thinking 2507 has a Design Arena: websites result, 1052.
Directional only
General
Vals Index
No call
Only Inkling Small has a Vals Index result, 27.6%.
Directional only
Long documents
Listed context window
Inkling Small
Inkling Small takes 524K tokens against 128K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Inkling Small
Qwen3 235B A22B Thinking 2507 $0.0022, Inkling Small $0.0010 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Inkling Small has the number, $0.16.
Directional only
At a glance
- Coding index
- Qwen3 235B A22B Thinking 2507 not benchmarked, Inkling Small 27.3 from 4 benchmarks.
- Widest category
- Agentic work, 22.2 points apart on average over 1 shared benchmark.
- Widest single gap
- AA Coding Index: Qwen3 235B A22B Thinking 2507 22.1 against Inkling Small 52.9, 30.8 points apart on a 0 to 100 scale.
- Output price
- Qwen3 235B A22B Thinking 2507 $3.50 against Inkling Small $1.20 per million output tokens, 2.9x apart.
- Context window
- Qwen3 235B A22B Thinking 2507 takes 128K tokens, Inkling Small 524K.
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 | Inkling Small | Ties |
|---|---|---|---|---|
| Coding | 0 | 3 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 3 | 2 | ||
| Math | 0 | 3 | ||
| Knowledge | 4 | 1 | ||
| General | 9 | 7 | ||
| All shared boards | 16 | 18 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index03
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| LMArena WebDev | Not measured | 1409 | Only Inkling Small measured |
| AA Coding Index | 22.1 | 52.9 | Inkling Small |
| LiveCodeBench (Vals AI) | Not measured | 85.9% | Only Inkling Small measured |
| SWE-bench Verified (Vals AI) | Not measured | 82.2% | Only Inkling Small measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 0.0% | Only Inkling Small measured |
| Vibe Code Bench (Vals AI) | Not measured | 19.1% | Only Inkling Small measured |
| Code migration (Vals AI) | Not measured | 13.7% | Only Inkling Small measured |
| ProgramBench (Vals AI) | Not measured | 0.5% | Only Inkling Small measured |
| IOI (Vals AI) | Not measured | 9.3% | Only Inkling Small measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 55.1% | Only Inkling Small measured |
| SkillsBench (Vals AI) | Not measured | 33.6% | Only Inkling Small measured |
| CyberBench (Vals AI) | Not measured | 64.1% | Only Inkling Small measured |
| SciCode | Not measured | 49.7% | Only Inkling Small measured |
| WeirdML | 41.0% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMArena Text: coding | 1424 | 1450 | Inkling Small |
| LMArena Text: software and IT | 1431 | 1447 | Inkling Small |
| LMArena WebDev: React | Not measured | 1399 | Only Inkling Small measured |
| LMArena WebDev: HTML | Not measured | 1398 | Only Inkling Small measured |
| LMArena WebDev: from an image | Not measured | 1410 | Only Inkling Small measured |
| Neither model has a result on the other 20 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| AA Intelligence Index | 12.7 | 25.7 | Inkling Small |
| Epoch Capabilities Index | Not measured | 150.3 | Only Inkling Small measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| AA Agentic Index | 1.3 | 23.5 | Inkling Small |
| Finance Agent (Vals AI) | Not measured | 41.3% | Only Inkling Small measured |
| Tax Agent Bench (Vals AI) | Not measured | 15.4% | Only Inkling Small measured |
| Legal Research Bench (Vals AI) | Not measured | 25.5% | Only Inkling Small measured |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 2.9% | Only Inkling Small measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 31.8% | Only Inkling Small measured |
| Vending-Bench 2 | -$11 | Not measured | Only 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 index32
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 83.6% | Only Inkling Small measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 89.5% | Only Inkling Small measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 33.3% | Only Inkling Small measured |
| GPQA Diamond (Epoch AI run) | 80.1% | 88.5% | Inkling Small |
| ARC-AGI | Not measured | 84.0% | Only Inkling Small measured |
| ARC-AGI-2 | Not measured | 40.1% | Only Inkling Small measured |
| Mystery game puzzles | 9.0% | 6.0% | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: hard prompts | 1416 | 1423 | Inkling Small |
| LMArena Text: expert | 1463 | 1441 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: hard prompts in English | 1432 | 1429 | Qwen3 235B A22B Thinking 2507 |
| LMArena Vision: diagrams | Not measured | 1251 | Only Inkling Small measured |
| LMArena Vision: homework | Not measured | 1257 | Only Inkling Small measured |
| Neither model has a result on the other 10 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| OTIS Mock AIME | 86.7% | 90.0% | Inkling Small |
| FrontierMath, tiers 1 to 3 | Not measured | 46.3% | Only Inkling Small measured |
| FrontierMath, tier 4 | Not measured | 17.1% | Only Inkling Small measured |
| ProofBench | Not measured | 6.0% | Only Inkling Small measured |
| LMArena Text: math | 1412 | 1459 | Inkling Small |
| LMArena Text: mathematical work | 1435 | 1441 | Inkling Small |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index41
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 85.6% | Only Inkling Small measured |
| LegalBench (Vals AI) | Not measured | 83.0% | Only Inkling Small measured |
| CorpFin (Vals AI) | Not measured | 69.6% | Only Inkling Small measured |
| TaxEval (Vals AI) | Not measured | 75.5% | Only Inkling Small measured |
| MortgageTax (Vals AI) | Not measured | 62.3% | Only Inkling Small measured |
| MedCode (Vals AI) | Not measured | 37.9% | Only Inkling Small measured |
| SimpleQA Verified | 40.4% | 19.1% | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: science | 1433 | 1428 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: medicine and healthcare | 1456 | 1413 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: legal and government | 1415 | 1415 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: business and finance | 1413 | 1413 | Inkling Small |
| Neither model has a result on the other 7 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| Design Arena: websites | 1052 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Design Arena: UI components | 942 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Design Arena: game dev | 956 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Design Arena: data viz | 950 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Design Arena: 3D | 997 | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| Design Arena: all code categories | 1032 | Not measured | Only 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 index97
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | Leader |
|---|---|---|---|
| Vals Index | Not measured | 27.6% | Only Inkling Small measured |
| Public Benefits Bench (Vals AI) | Not measured | 59.4% | Only Inkling Small measured |
| SAGE (Vals AI) | Not measured | 32.7% | Only Inkling Small measured |
| MedScribe (Vals AI) | Not measured | 84.1% | Only Inkling Small measured |
| Vals Multimodal Index | Not measured | 50.0% | Only Inkling Small measured |
| Long context reasoning (AA-LCR) | Not measured | 75.7% | Only Inkling Small measured |
| DTBench | 80.3% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| LMCA | 29.3% | Not measured | Only Qwen3 235B A22B Thinking 2507 measured |
| 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 | 1413 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: instruction following | 1386 | 1399 | Inkling Small |
| LMArena Text: creative writing | 1384 | 1334 | Qwen3 235B A22B Thinking 2507 |
| LMArena Vision | Not measured | 1237 | Only Inkling Small measured |
| LMArena Text: English | 1429 | 1421 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: non-English | 1397 | 1401 | Inkling Small |
| LMArena Text: Chinese | 1481 | 1464 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: German | 1408 | 1408 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: Japanese | 1386 | 1401 | Inkling Small |
| LMArena Text: Korean | 1375 | 1367 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: French | Not measured | 1435 | Only Inkling Small measured |
| LMArena Text: Spanish | 1389 | 1422 | Inkling Small |
| LMArena Text: Russian | 1401 | 1394 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: Polish | 1390 | 1428 | Inkling Small |
| LMArena Text: multi-turn | 1406 | 1416 | Inkling Small |
| LMArena Text: longer queries | 1400 | 1402 | Inkling Small |
| LMArena Text: writing and language | 1388 | 1357 | Qwen3 235B A22B Thinking 2507 |
| LMArena Text: entertainment and media | 1370 | 1352 | Qwen3 235B A22B Thinking 2507 |
| LMArena Vision: English | Not measured | 1243 | Only Inkling Small measured |
| LMArena Vision: Chinese | Not measured | 1292 | Only Inkling Small measured |
| LMArena Vision: reading text | Not measured | 1243 | Only Inkling Small measured |
| LMArena Vision: humor | Not measured | 1207 | Only Inkling Small measured |
| LMArena Vision: creative writing | Not measured | 1205 | Only Inkling Small measured |
| Neither model has a result on the other 13 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3 235B A22B Thinking 2507 | Inkling Small | 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
Inkling Small costs less
Repository review
50K input + 3K output tokens
Inkling Small costs less
Agent loop
200K input + 10K output tokens
Inkling Small 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 | Thinking Machines Inkling Small |
|---|---|---|
| Maker | Qwen | Thinking Machines |
| Weights | Open weights | Open weights |
| Released | July 25, 2025 | July 30, 2026 |
| Context window | 128K | 524K |
| Input $/Mtok | $0.45 | $0.45 |
| Output $/Mtok | $3.50 | $1.20 |
| Coding index | Not benchmarked | 27.3 |
| Independent measurements behind it | 0 | 4 |
| Cost per solved task | Not computable | $0.16 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3 235B A22B Thinking 2507 better than Inkling Small for coding?
Only Inkling Small has a coding score, 27.3 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 Mystery game puzzles or LMArena Text: expert, where it leads.
Which is cheaper, Qwen3 235B A22B Thinking 2507 or Inkling Small?
Qwen3 235B A22B Thinking 2507 costs $0.45 per million input tokens and $3.50 per million output tokens; Inkling Small costs $0.45 and $1.20. Cost per solved task is known for Inkling Small only.
Which has the bigger context window, Qwen3 235B A22B Thinking 2507 or Inkling Small?
Inkling Small, with 524K tokens against 128K.
How many benchmarks is the Qwen3 235B A22B Thinking 2507 vs Inkling Small comparison based on?
34 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 Inkling Small's on 4. Where only one model has a result, the page says so instead of naming a winner.