Qwen3.8 27B vs Inkling 2 Small
Qwen3.8 27B leads Inkling 2 Small on coding, 69.7 to 48.9 on the Vibe Coding Index, and Inkling 2 Small costs 2.5x less per output token ($1.00 vs $2.55 per million). Updated November 14, 2026; 39 benchmarks have a result for both.
Qwen3.8 27B or Inkling 2 Small: which to pick
Pick Qwen3.8 27B for coding: it scores 69.7 on the Vibe Coding Index against 48.9 for Inkling 2 Small.
Pick Inkling 2 Small for output tokens 2.5x cheaper ($1.00 vs $2.55 per million) and a larger context window (1.05M against 1.02M tokens).
- Shared results
- 39
- Qwen3.8 27B only
- 54
- Inkling 2 Small only
- 12
- Like-for-like categories
- 6 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
Qwen3.8 27B
Qwen3.8 27B scores 69.7 against 48.9.
4 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
Inkling 2 Small
Inkling 2 Small leads, 30.6 against 20.1.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Qwen3.8 27B has a AA Agentic Index result, 45.8.
Directional only
Reasoning
GPQA Diamond
No call
Only Qwen3.8 27B has a GPQA Diamond result, 88.9%.
Directional only
Knowledge
MMLU Pro
No call
Only Qwen3.8 27B has a MMLU Pro result, 84.3%.
Directional only
General
Vals Index
No call
Only Inkling 2 Small has a Vals Index result, 56.1%.
Directional only
Long documents
Listed context window
Inkling 2 Small
Inkling 2 Small takes 1.05M tokens against 1.02M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Inkling 2 Small
Qwen3.8 27B $0.0017, Inkling 2 Small $0.0008 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Qwen3.8 27B has the number, $1.26.
Directional only
At a glance
- Coding index
- Qwen3.8 27B 69.7 from 4 benchmarks, Inkling 2 Small 48.9 from 3 benchmarks.
- Widest category
- Reasoning, 12.8 points apart on average over 8 shared benchmarks.
- Widest single gap
- ARC-AGI-2: Qwen3.8 27B 42.4% against Inkling 2 Small 81.7%, 39.3 points apart on a 0 to 100 scale.
- Output price
- Qwen3.8 27B $2.55 against Inkling 2 Small $1.00 per million output tokens, 2.5x apart.
- Context window
- Qwen3.8 27B takes 1.02M tokens, Inkling 2 Small 1.05M.
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 | Qwen3.8 27B | Bars | Inkling 2 Small | Ties |
|---|---|---|---|---|
| Coding | 2 | 9 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 1 | 3 | ||
| Reasoning | 3 | 5 | ||
| Knowledge | 2 | 1 | ||
| General | 10 | 2 | ||
| All shared boards | 18 | 21 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index29
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| LiveBench coding | 69.3% | Not measured | Only Qwen3.8 27B measured |
| LMArena WebDev | 1590 | 1647 | Inkling 2 Small |
| AA Coding Index | 68.1 | Not measured | Only Qwen3.8 27B measured |
| LiveCodeBench (Vals AI) | 84.0% | Not measured | Only Qwen3.8 27B measured |
| SWE-bench Verified (Vals AI) | 86.0% | Not measured | Only Qwen3.8 27B measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 12.6% | Only Inkling 2 Small measured |
| Vibe Code Bench (Vals AI) | 64.8% | 81.5% | Inkling 2 Small |
| Code migration (Vals AI) | 14.2% | Not measured | Only Qwen3.8 27B measured |
| ProgramBench (Vals AI) | 0.0% | 0.5% | Inkling 2 Small |
| IOI (Vals AI) | 39.1% | Not measured | Only Qwen3.8 27B measured |
| Terminal-Bench 2.1 (Vals AI) | 58.4% | Not measured | Only Qwen3.8 27B measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 43.1% | Only Inkling 2 Small measured |
| SkillsBench (Vals AI) | 38.1% | Not measured | Only Qwen3.8 27B measured |
| CyberBench (Vals AI) | 67.8% | 72.7% | Inkling 2 Small |
| SciCode | 46.6% | 52.3% | Inkling 2 Small |
| Terminal-Bench 4.0 (Artificial Analysis run) | 5.6% | 13.1% | Inkling 2 Small |
| FrontierCode 2.0 | 14.9% | 25.4% | Inkling 2 Small |
| LMArena Text: coding | 1479 | 1455 | Qwen3.8 27B |
| LMArena Text: software and IT | 1474 | 1443 | Qwen3.8 27B |
| LMArena WebDev: React | 1596 | 1660 | Inkling 2 Small |
| LMArena WebDev: HTML | 1570 | 1661 | Inkling 2 Small |
| LMArena WebDev: from an image | 1574 | Not measured | Only Qwen3.8 27B measured |
| SWE-Bench Pro V2 | Not measured | 97.7% | Only Inkling 2 Small measured |
| GameBench 2 | 43.8 | Not measured | Only Qwen3.8 27B measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| AA Intelligence Index | 20.1 | 30.6 | Inkling 2 Small |
| Epoch Capabilities Index | 152.6 | Not measured | Only Qwen3.8 27B measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index13
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| AA Agentic Index | 45.8 | Not measured | Only Qwen3.8 27B measured |
| LiveBench research | 61.3% | Not measured | Only Qwen3.8 27B measured |
| LiveBench browser | 24.2% | Not measured | Only Qwen3.8 27B measured |
| LiveBench long agentic | 15.9% | Not measured | Only Qwen3.8 27B measured |
| SREBench (Vals AI) | Not measured | 1.9% | Only Inkling 2 Small measured |
| Finance Agent (Vals AI) | 48.5% | Not measured | Only Qwen3.8 27B measured |
| Tax Agent Bench (Vals AI) | 30.9% | 38.9% | Inkling 2 Small |
| Legal Research Bench (Vals AI) | 36.1% | 46.6% | Inkling 2 Small |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 21.7% | Only Inkling 2 Small measured |
| Excel Modeling Benchmark (Vals AI) | 59.7% | Not measured | Only Qwen3.8 27B measured |
| BioMysteryBench (Vals AI) | Not measured | 70.0% | Only Inkling 2 Small measured |
| Terminal-Bench-Science (Vals AI) | 0.0% | 2.9% | Inkling 2 Small |
| Vals Science Index | Not measured | 32.7% | Only Inkling 2 Small measured |
| APEX-Agents | 47.5% | 46.8% | Qwen3.8 27B |
| MCP Atlas | Not measured | 85.4% | Only Inkling 2 Small measured |
| Neither model has a result on the other 29 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index35
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| GPQA Diamond | 88.9% | Not measured | Only Qwen3.8 27B measured |
| LiveBench reasoning | 80.0% | Not measured | Only Qwen3.8 27B measured |
| MysteryMechanism (Vals AI) | Not measured | 25.2% | Only Inkling 2 Small measured |
| MMMU Pro (Vals AI) | 83.9% | Not measured | Only Qwen3.8 27B measured |
| GPQA Diamond (Artificial Analysis run) | 90.5% | Not measured | Only Qwen3.8 27B measured |
| Humanity's Last Exam (Artificial Analysis run) | 33.9% | 44.1% | Inkling 2 Small |
| HLE-Diamond (Artificial Analysis run) | 15.3% | 28.7% | Inkling 2 Small |
| CritPt v2, hard subset (Artificial Analysis run) | 16.5% | 35.0% | Inkling 2 Small |
| ARC-AGI | 87.5% | 96.5% | Inkling 2 Small |
| ARC-AGI-2 | 42.4% | 81.7% | Inkling 2 Small |
| SimpleBench | 60.2% | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: hard prompts | 1456 | 1421 | Qwen3.8 27B |
| LMArena Text: expert | 1479 | 1478 | Qwen3.8 27B |
| LMArena Text: hard prompts in English | 1462 | 1424 | Qwen3.8 27B |
| LMArena Vision: diagrams | 1281 | Not measured | Only Qwen3.8 27B measured |
| LMArena Vision: homework | 1284 | Not measured | Only Qwen3.8 27B measured |
| Neither model has a result on the other 10 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| ProofBench | 16.0% | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: math | 1455 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: mathematical work | 1461 | Not measured | Only Qwen3.8 27B measured |
| MathArena | Not measured | 42.4% | Only Inkling 2 Small measured |
| Neither model has a result on the other 10 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index21
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| MMLU Pro | 84.3% | Not measured | Only Qwen3.8 27B measured |
| LegalBench (Vals AI) | 82.4% | Not measured | Only Qwen3.8 27B measured |
| TaxEval (Vals AI) | 70.8% | Not measured | Only Qwen3.8 27B measured |
| MortgageTax (Vals AI) | 64.9% | Not measured | Only Qwen3.8 27B measured |
| MedCode (Vals AI) | 28.7% | Not measured | Only Qwen3.8 27B measured |
| AA-Omniscience | -6.4 | 16.1 | Inkling 2 Small |
| GeoBench | 38.0% | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: science | 1463 | 1409 | Qwen3.8 27B |
| LMArena Text: medicine and healthcare | 1439 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: legal and government | 1446 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: business and finance | 1440 | 1415 | Qwen3.8 27B |
| Neither model has a result on the other 10 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| LiveBench design | 72.4% | Not measured | Only Qwen3.8 27B measured |
| Neither model has a result on the other 8 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index102
| Benchmark | Qwen3.8 27B | Inkling 2 Small | Leader |
|---|---|---|---|
| Vals Index | Not measured | 56.1% | Only Inkling 2 Small measured |
| LiveBench data analysis | 76.6% | Not measured | Only Qwen3.8 27B measured |
| SAGE (Vals AI) | Not measured | 88.6% | Only Inkling 2 Small measured |
| MedScribe (Vals AI) | 83.8% | Not measured | Only Qwen3.8 27B measured |
| Long context reasoning (AA-LCR) | 82.0% | 83.3% | Inkling 2 Small |
| GDPval-AA | 1426 | 1630 | Inkling 2 Small |
| DTBench | 88.0% | Not measured | Only Qwen3.8 27B measured |
| LMCA | 41.4% | Not measured | Only Qwen3.8 27B measured |
| Surface Evolver Bench | 45.0% | Not measured | Only Qwen3.8 27B measured |
| LMArena Text | 1437 | 1403 | Qwen3.8 27B |
| LMArena Text: instruction following | 1431 | 1408 | Qwen3.8 27B |
| LMArena Text: creative writing | 1377 | 1348 | Qwen3.8 27B |
| LMArena Vision | 1270 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: English | 1447 | 1418 | Qwen3.8 27B |
| LMArena Text: non-English | 1424 | 1390 | Qwen3.8 27B |
| LMArena Text: Chinese | 1493 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: German | 1436 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: Japanese | 1398 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: Korean | 1389 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: French | 1466 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: Spanish | 1443 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: Russian | 1408 | 1377 | Qwen3.8 27B |
| LMArena Text: Polish | 1441 | Not measured | Only Qwen3.8 27B measured |
| LMArena Text: multi-turn | 1436 | 1408 | Qwen3.8 27B |
| LMArena Text: longer queries | 1446 | 1403 | Qwen3.8 27B |
| LMArena Text: writing and language | 1401 | 1381 | Qwen3.8 27B |
| LMArena Text: entertainment and media | 1388 | 1359 | Qwen3.8 27B |
| LMArena Vision: English | 1272 | Not measured | Only Qwen3.8 27B measured |
| LMArena Vision: Chinese | 1327 | Not measured | Only Qwen3.8 27B measured |
| LMArena Vision: reading text | 1280 | Not measured | Only Qwen3.8 27B measured |
| LMArena Vision: humor | 1245 | Not measured | Only Qwen3.8 27B measured |
| LMArena Vision: creative writing | 1234 | Not measured | Only Qwen3.8 27B measured |
| Neither model has a result on the other 17 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Qwen3.8 27B | Inkling 2 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 2 Small costs less
Repository review
50K input + 3K output tokens
Inkling 2 Small costs less
Agent loop
200K input + 10K output tokens
Inkling 2 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 | Qwen3.8 27B | Thinking Machines Inkling 2 Small |
|---|---|---|
| Maker | Qwen | Thinking Machines |
| Weights | Open weights | Open weights |
| Released | August 14, 2026 | November 9, 2026 |
| Context window | 1.02M | 1.05M |
| Input $/Mtok | $0.425 | $0.30 |
| Output $/Mtok | $2.55 | $1.00 |
| Coding index | 69.7 | 48.9 |
| Independent measurements behind it | 4 | 3 |
| Cost per solved task | $1.26 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Qwen3.8 27B better than Inkling 2 Small for coding?
Pick Qwen3.8 27B for coding: it scores 69.7 on the Vibe Coding Index against 48.9 for Inkling 2 Small. Pick Inkling 2 Small for output tokens 2.5x cheaper ($1.00 vs $2.55 per million) and a larger context window (1.05M against 1.02M tokens).
Which is cheaper, Qwen3.8 27B or Inkling 2 Small?
Qwen3.8 27B costs $0.425 per million input tokens and $2.55 per million output tokens; Inkling 2 Small costs $0.30 and $1.00. Cost per solved task is known for Qwen3.8 27B only.
Which has the bigger context window, Qwen3.8 27B or Inkling 2 Small?
Inkling 2 Small, with 1.05M tokens against 1.02M.
How many benchmarks is the Qwen3.8 27B vs Inkling 2 Small comparison based on?
39 benchmarks have a result for both models as of November 14, 2026. Qwen3.8 27B's coding index rests on 4 independent benchmarks and Inkling 2 Small's on 3. Where only one model has a result, the page says so instead of naming a winner.