Enter opens the highlighted result. Esc closes.

Inkling vs Inkling 2 Mini

Inkling 2 Mini leads Inkling on coding, 39.5 to 29.2 on the Vibe Coding Index, and Inkling 2 Mini costs 12.5x less per output token ($0.325 vs $4.05 per million). Updated November 22, 2026; 23 benchmarks have a result for both.

T InklingThinking Machines
Coding
29.2
Overall
42.3
Output $/M
$4.05
6
T Inkling 2 MiniThinking Machines
Coding
39.5
Overall
No data
Output $/M
$0.325
16
vs boards won of 23 shared, 1 tied

Inkling or Inkling 2 Mini: which to pick

Pick Inkling 2 Mini for coding: it scores 39.5 on the Vibe Coding Index against 29.2 for Inkling, though on 2 benchmarks against 6.

Pick Inkling for work measured by IOI (Vals AI) or Tax Agent Bench (Vals AI), where it leads.

Shared results
23
Inkling only
90
Inkling 2 Mini only
6
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.

At a glance

Coding index
Inkling 29.2 from 6 benchmarks, Inkling 2 Mini 39.5 from 2 benchmarks.
Widest category
Web design, 18.5 points apart on average over 3 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): Inkling 19.2% against Inkling 2 Mini 77.9%, 58.7 points apart on a 0 to 100 scale.
Output price
Inkling $4.05 against Inkling 2 Mini $0.325 per million output tokens, 12.5x apart.
Context window
Inkling takes 524K tokens, Inkling 2 Mini 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.

Boards won, where both have a result
CategoryInklingBarsInkling 2 MiniTies
Coding17
Intelligence01
Agentic work211 tie
Reasoning11
Knowledge20
Web design03
General03
All shared boards6161 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index17
Coding benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
LiveBench coding48.7%Not measuredOnly Inkling measured
LMArena WebDev1414Not measuredOnly Inkling measured
LiveCodeBench (Vals AI)85.5%Not measuredOnly Inkling measured
SWE-bench Verified (Vals AI)77.6%Not measuredOnly Inkling measured
Terminal-Bench 4 (Vals AI)0.5%6.6%Inkling 2 Mini
Vibe Code Bench (Vals AI)19.2%77.9%Inkling 2 Mini
Code migration (Vals AI)11.8%Not measuredOnly Inkling measured
ProgramBench (Vals AI)0.0%0.5%Inkling 2 Mini
IOI (Vals AI)14.9%11.6%Inkling
Terminal-Bench 2.1 (Vals AI)47.6%Not measuredOnly Inkling measured
SWE-bench Multimodal (Vals AI)Not measured37.9%Only Inkling 2 Mini measured
Vibe Code Bench 1-100 (Vals AI)7.3%Not measuredOnly Inkling measured
SkillsBench (Vals AI)26.1%43.2%Inkling 2 Mini
CyberBench (Vals AI)66.8%74.6%Inkling 2 Mini
SciCode47.0%Not measuredOnly Inkling measured
Terminal-Bench 4.0 (Artificial Analysis run)1.0%5.6%Inkling 2 Mini
FrontierCode 2.03.9%15.1%Inkling 2 Mini
FrontierCode14.0%Not measuredOnly Inkling measured
FrontierSWE4.1%Not measuredOnly Inkling measured
WeirdML32.3%Not measuredOnly Inkling measured
ALE-Bench946.0Not measuredOnly Inkling measured
LMArena Text: coding1467Not measuredOnly Inkling measured
LMArena Text: software and IT1466Not measuredOnly Inkling measured
LMArena WebDev: React1404Not measuredOnly Inkling measured
LMArena WebDev: HTML1402Not measuredOnly Inkling measured
SWE-Bench Pro V289.9%Not measuredOnly Inkling measured
GameBench 231.5Not measuredOnly Inkling measured
Terminal-Bench 4.00.0%Not measuredOnly Inkling measured
Neither model has a result on the other 24 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
AA Intelligence Index12.520.1Inkling 2 Mini
Epoch Capabilities Index148.4Not measuredOnly Inkling measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index21
Agentic work benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
Design Arena: agents building web apps1164Not measuredOnly Inkling measured
LiveBench research60.3%Not measuredOnly Inkling measured
LiveBench browser12.0%Not measuredOnly Inkling measured
LiveBench long agentic9.8%Not measuredOnly Inkling measured
SREBench (Vals AI)0.0%0.0%Tie
Finance Agent (Vals AI)46.6%Not measuredOnly Inkling measured
Tax Agent Bench (Vals AI)15.3%14.1%Inkling
Legal Research Bench (Vals AI)28.4%17.8%Inkling
Harvey's Legal Agent Benchmark (Vals AI)3.8%5.0%Inkling 2 Mini
Excel Modeling Benchmark (Vals AI)38.4%Not measuredOnly Inkling measured
BioMysteryBench (Vals AI)Not measured47.8%Only Inkling 2 Mini measured
Terminal-Bench-Science (Vals AI)Not measured0.0%Only Inkling 2 Mini measured
Vals Science IndexNot measured23.6%Only Inkling 2 Mini measured
APEX-Agents33.8%Not measuredOnly Inkling measured
MCP Atlas76.0%Not measuredOnly Inkling measured
Toolathlon-Verified45.5%Not measuredOnly Inkling measured
τ³-Banking25.0%Not measuredOnly Inkling measured
Neither model has a result on the other 27 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index11
Reasoning benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
GPQA Diamond87.1%Not measuredOnly Inkling measured
LiveBench reasoning78.3%Not measuredOnly Inkling measured
MysteryMechanism (Vals AI)Not measured23.0%Only Inkling 2 Mini measured
MMMU Pro (Vals AI)78.5%Not measuredOnly Inkling measured
GPQA Diamond (Artificial Analysis run)87.2%Not measuredOnly Inkling measured
Humanity's Last Exam (Artificial Analysis run)31.9%Not measuredOnly Inkling measured
HLE-Diamond (Artificial Analysis run)10.6%15.2%Inkling 2 Mini
CritPt v2, hard subset (Artificial Analysis run)18.0%17.0%Inkling
GPQA Diamond (Epoch AI run)88.3%Not measuredOnly Inkling measured
ARC-AGI79.5%Not measuredOnly Inkling measured
ARC-AGI-236.5%Not measuredOnly Inkling measured
SimpleBench50.0%Not measuredOnly Inkling measured
Chess puzzles21.0%Not measuredOnly Inkling measured
LMArena Text: hard prompts1453Not measuredOnly Inkling measured
LMArena Text: expert1469Not measuredOnly Inkling measured
LMArena Text: hard prompts in English1451Not measuredOnly Inkling measured
Humanity's Last Exam (Scale run)11.2%Not measuredOnly Inkling measured
Neither model has a result on the other 9 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Math benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
OTIS Mock AIME88.9%Not measuredOnly Inkling measured
FrontierMath, tiers 1 to 333.3%Not measuredOnly Inkling measured
FrontierMath, tier 44.9%Not measuredOnly Inkling measured
ProofBench0.0%Not measuredOnly Inkling measured
LMArena Text: math1481Not measuredOnly Inkling measured
LMArena Text: mathematical work1466Not measuredOnly Inkling measured
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index20
Knowledge benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
MMLU Pro86.3%Not measuredOnly Inkling measured
LegalBench (Vals AI)83.0%Not measuredOnly Inkling measured
CorpFin (Vals AI)68.6%Not measuredOnly Inkling measured
TaxEval (Vals AI)75.3%Not measuredOnly Inkling measured
MortgageTax (Vals AI)64.0%Not measuredOnly Inkling measured
MedCode (Vals AI)41.2%25.7%Inkling
AA-Omniscience-1.5-8.6Inkling
SimpleQA Verified40.3%Not measuredOnly Inkling measured
LMArena Text: science1461Not measuredOnly Inkling measured
LMArena Text: medicine and healthcare1451Not measuredOnly Inkling measured
LMArena Text: legal and government1450Not measuredOnly Inkling measured
LMArena Text: business and finance1439Not measuredOnly Inkling measured
Neither model has a result on the other 9 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index03
Web design benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
Design Arena: websites1210Not measuredOnly Inkling measured
Design Arena: UI components11821292Inkling 2 Mini
Design Arena: game dev11341287Inkling 2 Mini
Design Arena: data viz1158Not measuredOnly Inkling measured
Design Arena: 3D11261308Inkling 2 Mini
Design Arena: all code categories1186Not measuredOnly Inkling measured
Design Arena: SVG1095Not measuredOnly Inkling measured
Design Arena: ASCII art1082Not measuredOnly Inkling measured
LiveBench design55.2%Not measuredOnly Inkling measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index03
General benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
Vals Index29.0%46.5%Inkling 2 Mini
LiveBench data analysis72.8%Not measuredOnly Inkling measured
Public Benefits Bench (Vals AI)58.5%Not measuredOnly Inkling measured
SAGE (Vals AI)Not measured85.9%Only Inkling 2 Mini measured
MedScribe (Vals AI)85.4%86.4%Inkling 2 Mini
Vals Multimodal Index49.4%Not measuredOnly Inkling measured
Long context reasoning (AA-LCR)77.3%Not measuredOnly Inkling measured
GDPval-AA10801319Inkling 2 Mini
DTBench87.5%Not measuredOnly Inkling measured
LMCA37.6%Not measuredOnly Inkling measured
LMArena Text1443Not measuredOnly Inkling measured
LMArena Text: instruction following1427Not measuredOnly Inkling measured
LMArena Text: creative writing1395Not measuredOnly Inkling measured
LMArena Text: English1444Not measuredOnly Inkling measured
LMArena Text: non-English1435Not measuredOnly Inkling measured
LMArena Text: Chinese1494Not measuredOnly Inkling measured
LMArena Text: German1439Not measuredOnly Inkling measured
LMArena Text: Japanese1441Not measuredOnly Inkling measured
LMArena Text: Korean1412Not measuredOnly Inkling measured
LMArena Text: French1464Not measuredOnly Inkling measured
LMArena Text: Spanish1454Not measuredOnly Inkling measured
LMArena Text: Russian1434Not measuredOnly Inkling measured
LMArena Text: Polish1452Not measuredOnly Inkling measured
LMArena Text: multi-turn1440Not measuredOnly Inkling measured
LMArena Text: longer queries1437Not measuredOnly Inkling measured
LMArena Text: writing and language1411Not measuredOnly Inkling measured
LMArena Text: entertainment and media1398Not measuredOnly Inkling measured
EQ-Bench 41212Not measuredOnly Inkling measured
Neither model has a result on the other 21 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Inkling and Inkling 2 Mini
BenchmarkInklingInkling 2 MiniLeader
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$0.0030Fits in one request
Inkling 2 Mini$0.0002Fits in one request

Inkling 2 Mini costs less

Repository review

50K input + 3K output tokens

Inkling$0.062Fits in one request
Inkling 2 Mini$0.0049Fits in one request

Inkling 2 Mini costs less

Agent loop

200K input + 10K output tokens

Inkling$0.240Fits in one request
Inkling 2 Mini$0.019Fits in one request

Inkling 2 Mini 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 Inkling and Inkling 2 Mini
SpecificationThinking Machines InklingThinking Machines Inkling 2 Mini
MakerThinking MachinesThinking Machines
WeightsOpen weightsOpen weights
ReleasedJuly 17, 2026November 21, 2026
Context window524K1.05M
Input $/Mtok$1.00$0.0775
Output $/Mtok$4.05$0.325
Coding index29.239.5
Independent measurements behind it62
Cost per solved task$0.76Not computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Inkling better than Inkling 2 Mini for coding?

Pick Inkling 2 Mini for coding: it scores 39.5 on the Vibe Coding Index against 29.2 for Inkling, though on 2 benchmarks against 6. Pick Inkling for work measured by IOI (Vals AI) or Tax Agent Bench (Vals AI), where it leads.

Which is cheaper, Inkling or Inkling 2 Mini?

Inkling costs $1.00 per million input tokens and $4.05 per million output tokens; Inkling 2 Mini costs $0.0775 and $0.325. Cost per solved task is known for Inkling only.

Which has the bigger context window, Inkling or Inkling 2 Mini?

Inkling 2 Mini, with 1.05M tokens against 524K.

How many benchmarks is the Inkling vs Inkling 2 Mini comparison based on?

23 benchmarks have a result for both models as of November 22, 2026. Inkling's coding index rests on 6 independent benchmarks and Inkling 2 Mini's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next

Embed this comparison on your site

Free under CC BY 4.0, while the source link stays visible. Preview the frame · Terms and other embeds

<iframe src="https://llmbenchmark.ai/embed/compare/thinkingmachines-inkling-vs-thinkingmachines-inkling-2-mini" title="Inkling vs Inkling 2 Mini, LLMBenchmark.ai" width="100%" height="360" style="border:0;max-width:100%" loading="lazy"></iframe>
<p style="font:13px/1.4 sans-serif;margin:6px 0 0">Source: <a href="https://llmbenchmark.ai/compare/thinkingmachines-inkling-vs-thinkingmachines-inkling-2-mini">LLMBenchmark.ai, Inkling vs Inkling 2 Mini</a></p>