DeepSeek V3.2 vs Inkling 2 Small
Inkling 2 Small leads DeepSeek V3.2 on coding, 48.9 to 30.0 on the Vibe Coding Index, and DeepSeek V3.2 costs 2.5x less per output token ($0.40 vs $1.00 per million). Updated November 14, 2026; 28 benchmarks have a result for both.
DeepSeek V3.2 or Inkling 2 Small: which to pick
Pick Inkling 2 Small for coding: it scores 48.9 on the Vibe Coding Index against 30.0 for DeepSeek V3.2.
Pick DeepSeek V3.2 for output tokens 2.5x cheaper ($0.40 vs $1.00 per million) and work measured by LMArena Text: software and IT or LMArena Text: hard prompts, where it leads.
- Shared results
- 28
- DeepSeek V3.2 only
- 48
- Inkling 2 Small only
- 23
- 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
Inkling 2 Small
Inkling 2 Small scores 48.9 against 30.0.
3 benchmarks behind each score
Intelligence
AA Intelligence Index
No call
Only Inkling 2 Small has a AA Intelligence Index result, 30.6.
Directional only
Reasoning
GPQA Diamond
No call
Only DeepSeek V3.2 has a GPQA Diamond result, 80.3%.
Directional only
Math
AIME
No call
Only DeepSeek V3.2 has a AIME result, 84.6%.
Directional only
Knowledge
MMLU Pro
No call
Only DeepSeek V3.2 has a MMLU Pro result, 84.9%.
Directional only
Web design
Design Arena: websites
No call
Only DeepSeek V3.2 has a Design Arena: websites result, 1167.
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 164K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V3.2
DeepSeek V3.2 $0.0005, Inkling 2 Small $0.0008 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only DeepSeek V3.2 has the number, $0.99.
Directional only
At a glance
- Coding index
- DeepSeek V3.2 30.0 from 3 benchmarks, Inkling 2 Small 48.9 from 3 benchmarks.
- Widest category
- Coding, 33.1 points apart on average over 6 shared benchmarks.
- Widest single gap
- ARC-AGI-2: DeepSeek V3.2 4.0% against Inkling 2 Small 81.7%, 77.6 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V3.2 $0.40 against Inkling 2 Small $1.00 per million output tokens, 2.5x apart.
- Context window
- DeepSeek V3.2 takes 164K 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 | DeepSeek V3.2 | Bars | Inkling 2 Small | Ties |
|---|---|---|---|---|
| Coding | 1 | 5 | ||
| Agentic work | 0 | 1 | ||
| Reasoning | 2 | 4 | ||
| Math | 0 | 1 | ||
| Knowledge | 1 | 1 | ||
| General | 10 | 2 | ||
| All shared boards | 14 | 14 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index15
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| SWE-bench Verified | 70.0% | Not measured | Only DeepSeek V3.2 measured |
| SWE-bench Multilingual | 59.0% | Not measured | Only DeepSeek V3.2 measured |
| LMArena WebDev | 1362 | 1647 | Inkling 2 Small |
| AA Coding Index | 44.2 | Not measured | Only DeepSeek V3.2 measured |
| LiveCodeBench (Vals AI) | 80.7% | Not measured | Only DeepSeek V3.2 measured |
| SWE-bench Verified (Vals AI) | 67.6% | Not measured | Only DeepSeek V3.2 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 12.6% | Only Inkling 2 Small measured |
| Vibe Code Bench (Vals AI) | 5.1% | 81.5% | Inkling 2 Small |
| ProgramBench (Vals AI) | Not measured | 0.5% | Only Inkling 2 Small measured |
| SWE-bench Multimodal (Vals AI) | Not measured | 43.1% | Only Inkling 2 Small measured |
| CyberBench (Vals AI) | Not measured | 72.7% | Only Inkling 2 Small measured |
| LiveCodeBench (Artificial Analysis run) | 86.2% | Not measured | Only DeepSeek V3.2 measured |
| SciCode | Not measured | 52.3% | Only Inkling 2 Small measured |
| Terminal-Bench Hard (Artificial Analysis run) | 35.6% | Not measured | Only DeepSeek V3.2 measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | Not measured | 13.1% | Only Inkling 2 Small measured |
| FrontierCode 2.0 | Not measured | 25.4% | Only Inkling 2 Small measured |
| Terminal-Bench | 39.6% | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: coding | 1453 | 1455 | Inkling 2 Small |
| LMArena Text: software and IT | 1450 | 1443 | DeepSeek V3.2 |
| LMArena WebDev: React | 1350 | 1660 | Inkling 2 Small |
| LMArena WebDev: HTML | 1370 | 1661 | Inkling 2 Small |
| SWE-Bench Pro V2 | Not measured | 97.7% | Only Inkling 2 Small measured |
| Terminal-Bench 2.0 | 39.6% | Not measured | Only DeepSeek V3.2 measured |
| Neither model has a result on the other 29 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 30.6 | Only Inkling 2 Small measured |
| Epoch Capabilities Index | 146.3 | Not measured | Only DeepSeek V3.2 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| SREBench (Vals AI) | Not measured | 1.9% | Only Inkling 2 Small measured |
| Tax Agent Bench (Vals AI) | Not measured | 38.9% | Only Inkling 2 Small measured |
| Legal Research Bench (Vals AI) | Not measured | 46.6% | Only Inkling 2 Small measured |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 21.7% | Only Inkling 2 Small measured |
| BioMysteryBench (Vals AI) | Not measured | 70.0% | Only Inkling 2 Small measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 2.9% | Only Inkling 2 Small measured |
| Vals Science Index | Not measured | 32.7% | Only Inkling 2 Small measured |
| τ²-Bench Telecom | 90.6% | Not measured | Only DeepSeek V3.2 measured |
| APEX-Agents | 21.3% | 46.8% | Inkling 2 Small |
| Vending-Bench 2 | $1,034 | Not measured | Only DeepSeek V3.2 measured |
| MCP Atlas | Not measured | 85.4% | Only Inkling 2 Small measured |
| Neither model has a result on the other 33 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index24
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| GPQA Diamond | 80.3% | Not measured | Only DeepSeek V3.2 measured |
| MysteryMechanism (Vals AI) | Not measured | 25.2% | Only Inkling 2 Small measured |
| GPQA Diamond (Artificial Analysis run) | 84.0% | Not measured | Only DeepSeek V3.2 measured |
| Humanity's Last Exam (Artificial Analysis run) | 24.6% | 44.1% | Inkling 2 Small |
| HLE-Diamond (Artificial Analysis run) | Not measured | 28.7% | Only Inkling 2 Small measured |
| CritPt v2, hard subset (Artificial Analysis run) | Not measured | 35.0% | Only Inkling 2 Small measured |
| ARC-AGI | 57.0% | 96.5% | Inkling 2 Small |
| ARC-AGI-2 | 4.0% | 81.7% | Inkling 2 Small |
| LMArena Text: hard prompts | 1434 | 1421 | DeepSeek V3.2 |
| LMArena Text: expert | 1435 | 1478 | Inkling 2 Small |
| LMArena Text: hard prompts in English | 1446 | 1424 | DeepSeek V3.2 |
| Neither model has a result on the other 15 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| AIME | 84.6% | Not measured | Only DeepSeek V3.2 measured |
| MGSM (Vals AI) | 86.0% | Not measured | Only DeepSeek V3.2 measured |
| AIME 2025 (Artificial Analysis run) | 92.0% | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: math | 1435 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: mathematical work | 1439 | Not measured | Only DeepSeek V3.2 measured |
| MathArena | 15.9% | 42.4% | Inkling 2 Small |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| MMLU Pro | 84.9% | Not measured | Only DeepSeek V3.2 measured |
| LegalBench (Vals AI) | 76.1% | Not measured | Only DeepSeek V3.2 measured |
| CaseLaw (Vals AI) | 55.4% | Not measured | Only DeepSeek V3.2 measured |
| CorpFin (Vals AI) | 51.0% | Not measured | Only DeepSeek V3.2 measured |
| TaxEval (Vals AI) | 68.2% | Not measured | Only DeepSeek V3.2 measured |
| MedQA (Vals AI) | 93.9% | Not measured | Only DeepSeek V3.2 measured |
| MMLU Pro (Artificial Analysis run) | 86.2% | Not measured | Only DeepSeek V3.2 measured |
| AA-Omniscience | Not measured | 16.1 | Only Inkling 2 Small measured |
| LMArena Text: science | 1446 | 1409 | DeepSeek V3.2 |
| LMArena Text: medicine and healthcare | 1441 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: legal and government | 1425 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: business and finance | 1410 | 1415 | Inkling 2 Small |
| Neither model has a result on the other 9 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| Design Arena: websites | 1167 | Not measured | Only DeepSeek V3.2 measured |
| Design Arena: UI components | 1139 | Not measured | Only DeepSeek V3.2 measured |
| Design Arena: game dev | 1121 | Not measured | Only DeepSeek V3.2 measured |
| Design Arena: data viz | 1155 | Not measured | Only DeepSeek V3.2 measured |
| Design Arena: 3D | 1112 | Not measured | Only DeepSeek V3.2 measured |
| Design Arena: all code categories | 1153 | Not measured | Only DeepSeek V3.2 measured |
| Design Arena: SVG | 1021 | Not measured | Only DeepSeek V3.2 measured |
| Design Arena: ASCII art | 1077 | Not measured | Only DeepSeek V3.2 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index102
| Benchmark | DeepSeek V3.2 | Inkling 2 Small | Leader |
|---|---|---|---|
| Vals Index | Not measured | 56.1% | Only Inkling 2 Small measured |
| SAGE (Vals AI) | Not measured | 88.6% | Only Inkling 2 Small measured |
| IFBench | 60.7% | Not measured | Only DeepSeek V3.2 measured |
| Long context reasoning (AA-LCR) | 73.3% | 83.3% | Inkling 2 Small |
| GDPval-AA | 699 | 1630 | Inkling 2 Small |
| DTBench | 85.6% | Not measured | Only DeepSeek V3.2 measured |
| LMCA | 28.8% | Not measured | Only DeepSeek V3.2 measured |
| CL-bench | 12.4% | Not measured | Only DeepSeek V3.2 measured |
| CL-bench Life | 7.4% | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text | 1425 | 1403 | DeepSeek V3.2 |
| LMArena Text: instruction following | 1413 | 1408 | DeepSeek V3.2 |
| LMArena Text: creative writing | 1400 | 1348 | DeepSeek V3.2 |
| LMArena Text: English | 1440 | 1418 | DeepSeek V3.2 |
| LMArena Text: non-English | 1409 | 1390 | DeepSeek V3.2 |
| LMArena Text: Chinese | 1455 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: German | 1412 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Japanese | 1377 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Korean | 1370 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: French | 1429 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Spanish | 1424 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Russian | 1424 | 1377 | DeepSeek V3.2 |
| LMArena Text: Polish | 1419 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: multi-turn | 1427 | 1408 | DeepSeek V3.2 |
| LMArena Text: longer queries | 1428 | 1403 | DeepSeek V3.2 |
| LMArena Text: writing and language | 1407 | 1381 | DeepSeek V3.2 |
| LMArena Text: entertainment and media | 1395 | 1359 | DeepSeek V3.2 |
| Neither model has a result on the other 23 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V3.2 | 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
DeepSeek V3.2 costs less
Repository review
50K input + 3K output tokens
DeepSeek V3.2 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V3.2 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 | DeepSeek V3.2 | Thinking Machines Inkling 2 Small |
|---|---|---|
| Maker | DeepSeek | Thinking Machines |
| Weights | Open weights | Open weights |
| Released | December 1, 2025 | November 9, 2026 |
| Context window | 164K | 1.05M |
| Input $/Mtok | $0.26 | $0.30 |
| Output $/Mtok | $0.40 | $1.00 |
| Coding index | 30.0 | 48.9 |
| Independent measurements behind it | 3 | 3 |
| Cost per solved task | $0.99 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V3.2 better than Inkling 2 Small for coding?
Pick Inkling 2 Small for coding: it scores 48.9 on the Vibe Coding Index against 30.0 for DeepSeek V3.2. Pick DeepSeek V3.2 for output tokens 2.5x cheaper ($0.40 vs $1.00 per million) and work measured by LMArena Text: software and IT or LMArena Text: hard prompts, where it leads.
Which is cheaper, DeepSeek V3.2 or Inkling 2 Small?
DeepSeek V3.2 costs $0.26 per million input tokens and $0.40 per million output tokens; Inkling 2 Small costs $0.30 and $1.00. Cost per solved task is known for DeepSeek V3.2 only.
Which has the bigger context window, DeepSeek V3.2 or Inkling 2 Small?
Inkling 2 Small, with 1.05M tokens against 164K.
How many benchmarks is the DeepSeek V3.2 vs Inkling 2 Small comparison based on?
28 benchmarks have a result for both models as of November 14, 2026. DeepSeek V3.2's coding index rests on 3 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.