Enter opens the highlighted result. Esc closes.

Gemini 4 Carbon vs Inkling Small

Gemini 4 Carbon leads Inkling Small on coding, 86.6 to 27.3 on the Vibe Coding Index, and Inkling Small costs 8.3x less per output token ($1.20 vs $10.00 per million). Updated November 14, 2026; 47 benchmarks have a result for both.

Replace Gemini 4 Carbon Replace Inkling Small
Gemini 4 CarbonGoogle
Coding
86.6
Overall
88.3
Output $/M
$10.00
47
T Inkling SmallThinking Machines
Coding
27.3
Overall
40.9
Output $/M
$1.20
0
vs boards won of 47 shared

Gemini 4 Carbon or Inkling Small: which to pick

Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 27.3 for Inkling Small.

Pick Inkling Small for output tokens 8.3x cheaper ($1.20 vs $10.00 per million).

Model AGemini 4 CarbonGoogle86.6/1004 independent benchmarksNo cost per solved taskModel BInkling SmallThinking Machines27.3/1004 independent benchmarks$0.16 per solved task
Shared results
47
Gemini 4 Carbon only
26
Inkling Small only
46
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
Gemini 4 Carbon 86.6 from 4 benchmarks, Inkling Small 27.3 from 4 benchmarks.
Widest category
Coding, 45.5 points apart on average over 9 shared benchmarks.
Widest single gap
Terminal-Bench 4 (Vals AI): Gemini 4 Carbon 63.1% against Inkling Small 0.0%, 84.2 points apart on a 0 to 100 scale.
Output price
Gemini 4 Carbon $10.00 against Inkling Small $1.20 per million output tokens, 8.3x apart.
Context window
Gemini 4 Carbon takes 1.05M 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.

Boards won, where both have a result
CategoryGemini 4 CarbonBarsInkling SmallTies
Coding90
Intelligence20
Agentic work30
Reasoning90
Math40
Knowledge60
General140
All shared boards470
CodingLike-for-likeSix of these boards feed the Vibe Coding Index90
Coding benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
LiveBench coding91.0%Not measuredOnly Gemini 4 Carbon measured
LMArena WebDev17681408Gemini 4 Carbon
AA Coding IndexNot measured52.9Only Inkling Small measured
LiveCodeBench (Vals AI)Not measured85.9%Only Inkling Small measured
SWE-bench Verified (Vals AI)Not measured82.2%Only Inkling Small measured
Terminal-Bench 4 (Vals AI)63.1%0.0%Gemini 4 Carbon
Vibe Code Bench (Vals AI)89.1%19.1%Gemini 4 Carbon
Code migration (Vals AI)Not measured13.7%Only Inkling Small measured
ProgramBench (Vals AI)Not measured0.5%Only Inkling Small measured
IOI (Vals AI)Not measured9.3%Only Inkling Small measured
Terminal-Bench 2.1 (Vals AI)Not measured55.1%Only Inkling Small measured
SkillsBench (Vals AI)Not measured33.6%Only Inkling Small measured
CyberBench (Vals AI)Not measured64.1%Only Inkling Small measured
SciCodeNot measured49.7%Only Inkling Small measured
Terminal-Bench 4.0 (Artificial Analysis run)65.2%1.0%Gemini 4 Carbon
FrontierCode 2.040.6%4.8%Gemini 4 Carbon
FrontierSWE61.2%Not measuredOnly Gemini 4 Carbon measured
ALE-Bench2374.7Not measuredOnly Gemini 4 Carbon measured
LMArena Text: coding15641451Gemini 4 Carbon
LMArena Text: software and IT15521448Gemini 4 Carbon
LMArena WebDev: React17921397Gemini 4 Carbon
LMArena WebDev: HTML17701396Gemini 4 Carbon
LMArena WebDev: from an imageNot measured1410Only Inkling Small measured
GameBench 2Not measured33.1Only Inkling Small measured
Terminal-Bench 4.0Not measured0.9%Only Inkling Small measured
CursorBench 4.060.4%Not measuredOnly Gemini 4 Carbon measured
Neither model has a result on the other 26 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
AA Intelligence Index60.012.9Gemini 4 Carbon
Epoch Capabilities Index169.8150.8Gemini 4 Carbon
Agentic workLike-for-likePublic boards, shown as published, kept out of the index30
Agentic work benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
AA Agentic IndexNot measured23.5Only Inkling Small measured
Design Arena: agents building mobile apps1284Not measuredOnly Gemini 4 Carbon measured
Design Arena: spreadsheets1224Not measuredOnly Gemini 4 Carbon measured
LiveBench research92.5%Not measuredOnly Gemini 4 Carbon measured
LiveBench browser80.0%Not measuredOnly Gemini 4 Carbon measured
LiveBench long agentic72.7%Not measuredOnly Gemini 4 Carbon measured
Finance Agent (Vals AI)Not measured41.3%Only Inkling Small measured
Tax Agent Bench (Vals AI)51.8%15.4%Gemini 4 Carbon
Legal Research Bench (Vals AI)55.8%25.5%Gemini 4 Carbon
Harvey's Legal Agent Benchmark (Vals AI)17.5%2.9%Gemini 4 Carbon
Excel Modeling Benchmark (Vals AI)Not measured31.8%Only Inkling Small measured
APEX-Agents84.2%Not measuredOnly Gemini 4 Carbon measured
Vending-Bench 2$16,484Not measuredOnly Gemini 4 Carbon measured
MCP AtlasNot measured79.2%Only Inkling Small measured
Toolathlon-VerifiedNot measured54.4%Only Inkling Small measured
Neither model has a result on the other 29 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index90
Reasoning benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
GPQA DiamondNot measured83.6%Only Inkling Small measured
LiveBench reasoning90.6%Not measuredOnly Gemini 4 Carbon measured
GPQA Diamond (Artificial Analysis run)Not measured89.5%Only Inkling Small measured
Humanity's Last Exam (Artificial Analysis run)Not measured33.3%Only Inkling Small measured
HLE-Diamond (Artificial Analysis run)69.1%12.1%Gemini 4 Carbon
CritPt v2, hard subset (Artificial Analysis run)69.0%18.5%Gemini 4 Carbon
GPQA Diamond (Epoch AI run)96.5%88.5%Gemini 4 Carbon
ARC-AGI98.5%84.0%Gemini 4 Carbon
ARC-AGI-294.2%40.1%Gemini 4 Carbon
SimpleBench92.6%Not measuredOnly Gemini 4 Carbon measured
Chess puzzles78.0%Not measuredOnly Gemini 4 Carbon measured
Mystery game puzzles73.0%6.0%Gemini 4 Carbon
EBR-bench87.6%Not measuredOnly Gemini 4 Carbon measured
LMArena Text: hard prompts15511425Gemini 4 Carbon
LMArena Text: expert15821440Gemini 4 Carbon
LMArena Text: hard prompts in English15451429Gemini 4 Carbon
LMArena Vision: diagramsNot measured1253Only Inkling Small measured
LMArena Vision: homeworkNot measured1252Only Inkling Small measured
Neither model has a result on the other 8 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index40
Math benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
OTIS Mock AIME100.0%90.0%Gemini 4 Carbon
FrontierMath, tiers 1 to 390.5%46.3%Gemini 4 Carbon
FrontierMath, tier 495.1%17.1%Gemini 4 Carbon
ProofBenchNot measured6.0%Only Inkling Small measured
LMArena Text: mathNot measured1459Only Inkling Small measured
LMArena Text: mathematical work15381437Gemini 4 Carbon
MathArena85.4%Not measuredOnly Gemini 4 Carbon measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index60
Knowledge benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
MMLU ProNot measured85.6%Only Inkling Small measured
LegalBench (Vals AI)Not measured83.0%Only Inkling Small measured
CorpFin (Vals AI)Not measured69.6%Only Inkling Small measured
TaxEval (Vals AI)Not measured75.5%Only Inkling Small measured
MortgageTax (Vals AI)Not measured62.3%Only Inkling Small measured
MedCode (Vals AI)Not measured37.9%Only Inkling Small measured
AA-Omniscience53.5-9.5Gemini 4 Carbon
SimpleQA Verified77.7%19.1%Gemini 4 Carbon
LMArena Text: science15351429Gemini 4 Carbon
LMArena Text: medicine and healthcare15251414Gemini 4 Carbon
LMArena Text: legal and government15421413Gemini 4 Carbon
LMArena Text: business and finance15201413Gemini 4 Carbon
FACTS Benchmark Suite78.4%Not measuredOnly Gemini 4 Carbon measured
PRBench Legal53.3%Not measuredOnly Gemini 4 Carbon measured
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
Web design benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
Design Arena: websites1319Not measuredOnly Gemini 4 Carbon measured
Design Arena: UI components1357Not measuredOnly Gemini 4 Carbon measured
Design Arena: game dev1326Not measuredOnly Gemini 4 Carbon measured
Design Arena: 3D1434Not measuredOnly Gemini 4 Carbon measured
Design Arena: all code categories1348Not measuredOnly Gemini 4 Carbon measured
LiveBench design80.2%Not measuredOnly Gemini 4 Carbon measured
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index140
General benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
Vals Index70.5%27.6%Gemini 4 Carbon
LiveBench data analysis86.5%Not measuredOnly Gemini 4 Carbon measured
Public Benefits Bench (Vals AI)Not measured59.4%Only Inkling Small measured
MedScribe (Vals AI)Not measured84.1%Only Inkling Small measured
Vals Multimodal IndexNot measured50.0%Only Inkling Small measured
Long context reasoning (AA-LCR)79.3%75.7%Gemini 4 Carbon
GDPval-AA17331133Gemini 4 Carbon
Furniture assembly77.5%Not measuredOnly Gemini 4 Carbon measured
LMArena Text15381414Gemini 4 Carbon
LMArena Text: instruction following15391402Gemini 4 Carbon
LMArena Text: creative writing15331337Gemini 4 Carbon
LMArena VisionNot measured1236Only Inkling Small measured
LMArena Text: English15301421Gemini 4 Carbon
LMArena Text: non-English15341402Gemini 4 Carbon
LMArena Text: Chinese16131462Gemini 4 Carbon
LMArena Text: GermanNot measured1403Only Inkling Small measured
LMArena Text: JapaneseNot measured1397Only Inkling Small measured
LMArena Text: KoreanNot measured1384Only Inkling Small measured
LMArena Text: FrenchNot measured1438Only Inkling Small measured
LMArena Text: SpanishNot measured1431Only Inkling Small measured
LMArena Text: Russian15601397Gemini 4 Carbon
LMArena Text: PolishNot measured1427Only Inkling Small measured
LMArena Text: multi-turn15691417Gemini 4 Carbon
LMArena Text: longer queries15491403Gemini 4 Carbon
LMArena Text: writing and language15421359Gemini 4 Carbon
LMArena Text: entertainment and media15271353Gemini 4 Carbon
LMArena Vision: EnglishNot measured1243Only Inkling Small measured
LMArena Vision: ChineseNot measured1291Only Inkling Small measured
LMArena Vision: reading textNot measured1242Only Inkling Small measured
LMArena Vision: humorNot measured1199Only Inkling Small measured
LMArena Vision: creative writingNot measured1197Only Inkling Small measured
Neither model has a result on the other 18 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 4 Carbon and Inkling Small
BenchmarkGemini 4 CarbonInkling SmallLeader
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

Gemini 4 Carbon$0.0070Fits in one request
Inkling Small$0.0010Fits in one request

Inkling Small costs less

Repository review

50K input + 3K output tokens

Gemini 4 Carbon$0.130Fits in one request
Inkling Small$0.026Fits in one request

Inkling Small costs less

Agent loop

200K input + 10K output tokens

Gemini 4 Carbon$0.500Fits in one request
Inkling Small$0.102Fits in one request

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.

Specifications of Gemini 4 Carbon and Inkling Small
SpecificationGoogle Gemini 4 CarbonThinking Machines Inkling Small
MakerGoogleThinking Machines
WeightsNo public weights listedOpen weights
ReleasedNovember 10, 2026July 30, 2026
Context window1.05M524K
Input $/Mtok$2.00$0.45
Output $/Mtok$10.00$1.20
Coding index86.627.3
Independent measurements behind it44
Cost per solved taskNot computable$0.16
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 4 Carbon better than Inkling Small for coding?

Pick Gemini 4 Carbon for coding: it scores 86.6 on the Vibe Coding Index against 27.3 for Inkling Small. Pick Inkling Small for output tokens 8.3x cheaper ($1.20 vs $10.00 per million).

Which is cheaper, Gemini 4 Carbon or Inkling Small?

Gemini 4 Carbon costs $2.00 per million input tokens and $10.00 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, Gemini 4 Carbon or Inkling Small?

Gemini 4 Carbon, with 1.05M tokens against 524K.

How many benchmarks is the Gemini 4 Carbon vs Inkling Small comparison based on?

47 benchmarks have a result for both models as of November 14, 2026. Gemini 4 Carbon's coding index rests on 4 independent benchmarks and Inkling Small's on 4. 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/gemini-4-carbon-vs-thinkingmachines-inkling-small" title="Gemini 4 Carbon vs Inkling Small, 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/gemini-4-carbon-vs-thinkingmachines-inkling-small">LLMBenchmark.ai, Gemini 4 Carbon vs Inkling Small</a></p>