Enter opens the highlighted result. Esc closes.

Gemini 4 Neon vs Inkling 2 Small

Gemini 4 Neon leads Inkling 2 Small on coding, 86.9 to 48.8 on the Vibe Coding Index, and Inkling 2 Small costs 15.0x less per output token ($1.00 vs $15.00 per million). Updated November 22, 2026; 38 benchmarks have a result for both.

Gemini 4 NeonGoogle
Coding
86.9
Overall
89.1
Output $/M
$15.00
37
T Inkling 2 SmallThinking Machines
Coding
48.8
Overall
56.9
Output $/M
$1.00
1
vs boards won of 38 shared

Gemini 4 Neon or Inkling 2 Small: which to pick

Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 48.8 for Inkling 2 Small.

Pick Inkling 2 Small for output tokens 15.0x cheaper ($1.00 vs $15.00 per million) and work measured by Harvey's Legal Agent Benchmark (Vals AI), where it leads.

Shared results
38
Gemini 4 Neon only
31
Inkling 2 Small only
17
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.

At a glance

Coding index
Gemini 4 Neon 86.9 from 4 benchmarks, Inkling 2 Small 48.8 from 3 benchmarks.
Widest category
Intelligence, 30.8 points apart on average over 1 shared benchmark.
Widest single gap
Terminal-Bench 4 (Vals AI): Gemini 4 Neon 61.6% against Inkling 2 Small 12.6%, 65.3 points apart on a 0 to 100 scale.
Output price
Gemini 4 Neon $15.00 against Inkling 2 Small $1.00 per million output tokens, 15.0x apart.
Context window
Both take 1.05M tokens.

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 NeonBarsInkling 2 SmallTies
Coding100
Intelligence10
Agentic work61
Reasoning60
Knowledge30
General110
All shared boards371
CodingLike-for-likeSix of these boards feed the Vibe Coding Index100
Coding benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
LiveBench coding89.7%Not measuredOnly Gemini 4 Neon measured
LMArena WebDev17651643Gemini 4 Neon
Terminal-Bench 4 (Vals AI)61.6%12.6%Gemini 4 Neon
Vibe Code Bench (Vals AI)93.7%81.5%Gemini 4 Neon
ProgramBench (Vals AI)Not measured0.5%Only Inkling 2 Small measured
SWE-bench Multimodal (Vals AI)Not measured43.1%Only Inkling 2 Small measured
CyberBench (Vals AI)Not measured72.7%Only Inkling 2 Small measured
SciCodeNot measured52.3%Only Inkling 2 Small measured
Terminal-Bench 4.0 (Artificial Analysis run)63.6%13.1%Gemini 4 Neon
FrontierCode 2.044.4%25.4%Gemini 4 Neon
FrontierSWENot measured25.1%Only Inkling 2 Small measured
LMArena Text: coding15551456Gemini 4 Neon
LMArena Text: software and IT15421444Gemini 4 Neon
LMArena WebDev: React17871655Gemini 4 Neon
LMArena WebDev: HTML18031629Gemini 4 Neon
SWE-Bench Pro V2Not measured97.7%Only Inkling 2 Small measured
GameBench 298.046.1Gemini 4 Neon
CursorBench 4.065.4%Not measuredOnly Gemini 4 Neon measured
Neither model has a result on the other 34 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
AA Intelligence Index61.430.6Gemini 4 Neon
Neither model has a result on the other 1 board here
Agentic workLike-for-likePublic boards, shown as published, kept out of the index61
Agentic work benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
LiveBench research94.5%Not measuredOnly Gemini 4 Neon measured
LiveBench browser95.6%Not measuredOnly Gemini 4 Neon measured
LiveBench long agentic82.6%Not measuredOnly Gemini 4 Neon measured
SREBench (Vals AI)Not measured1.9%Only Inkling 2 Small measured
Tax Agent Bench (Vals AI)54.9%38.9%Gemini 4 Neon
Legal Research Bench (Vals AI)56.3%46.6%Gemini 4 Neon
Harvey's Legal Agent Benchmark (Vals AI)19.6%21.7%Inkling 2 Small
BioMysteryBench (Vals AI)80.0%70.0%Gemini 4 Neon
Terminal-Bench-Science (Vals AI)62.9%2.9%Gemini 4 Neon
Vals Science Index62.6%32.7%Gemini 4 Neon
APEX-Agents82.9%46.8%Gemini 4 Neon
Vending-Bench 2$16,843Not measuredOnly Gemini 4 Neon measured
MCP AtlasNot measured85.4%Only Inkling 2 Small measured
Neither model has a result on the other 31 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
Reasoning benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
LiveBench reasoning89.6%Not measuredOnly Gemini 4 Neon measured
MysteryMechanism (Vals AI)45.0%25.2%Gemini 4 Neon
Humanity's Last Exam (Artificial Analysis run)Not measured44.1%Only Inkling 2 Small measured
HLE-Diamond (Artificial Analysis run)71.3%28.7%Gemini 4 Neon
CritPt v2, hard subset (Artificial Analysis run)61.5%35.0%Gemini 4 Neon
GPQA Diamond (Epoch AI run)96.3%Not measuredOnly Gemini 4 Neon measured
ARC-AGINot measured96.5%Only Inkling 2 Small measured
ARC-AGI-2Not measured81.7%Only Inkling 2 Small measured
SimpleBench94.4%Not measuredOnly Gemini 4 Neon measured
Chess puzzles83.0%Not measuredOnly Gemini 4 Neon measured
Mystery game puzzles85.0%Not measuredOnly Gemini 4 Neon measured
EBR-bench98.1%Not measuredOnly Gemini 4 Neon measured
LMArena Text: hard prompts15471421Gemini 4 Neon
LMArena Text: expert15601478Gemini 4 Neon
LMArena Text: hard prompts in English15241424Gemini 4 Neon
Neither model has a result on the other 11 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Math benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
OTIS Mock AIME100.0%Not measuredOnly Gemini 4 Neon measured
FrontierMath, tiers 1 to 391.2%Not measuredOnly Gemini 4 Neon measured
FrontierMath, tier 497.6%Not measuredOnly Gemini 4 Neon measured
LMArena Text: math1548Not measuredOnly Gemini 4 Neon measured
LMArena Text: mathematical work1566Not measuredOnly Gemini 4 Neon measured
MathArenaNot measured42.3%Only Inkling 2 Small measured
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index30
Knowledge benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
MedCode (Vals AI)Not measured44.1%Only Inkling 2 Small measured
AA-Omniscience54.816.1Gemini 4 Neon
SimpleQA Verified75.2%Not measuredOnly Gemini 4 Neon measured
LMArena Text: science15491409Gemini 4 Neon
LMArena Text: medicine and healthcare1534Not measuredOnly Gemini 4 Neon measured
LMArena Text: legal and government1518Not measuredOnly Gemini 4 Neon measured
LMArena Text: business and finance14931416Gemini 4 Neon
Neither model has a result on the other 14 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Web design benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
Design Arena: websites1347Not measuredOnly Gemini 4 Neon measured
Design Arena: UI components1349Not measuredOnly Gemini 4 Neon measured
Design Arena: game dev1394Not measuredOnly Gemini 4 Neon measured
Design Arena: 3D1490Not measuredOnly Gemini 4 Neon measured
Design Arena: all code categories1391Not measuredOnly Gemini 4 Neon measured
LiveBench design81.5%Not measuredOnly Gemini 4 Neon measured
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index110
General benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 SmallLeader
Vals Index70.9%56.1%Gemini 4 Neon
LiveBench data analysis85.3%Not measuredOnly Gemini 4 Neon measured
SAGE (Vals AI)Not measured88.6%Only Inkling 2 Small measured
MedScribe (Vals AI)Not measured90.2%Only Inkling 2 Small measured
Long context reasoning (AA-LCR)Not measured83.3%Only Inkling 2 Small measured
GDPval-AANot measured1630Only Inkling 2 Small measured
Furniture assembly89.2%Not measuredOnly Gemini 4 Neon measured
LMArena Text15241403Gemini 4 Neon
LMArena Text: instruction following15351407Gemini 4 Neon
LMArena Text: creative writing15191346Gemini 4 Neon
LMArena Text: English15111418Gemini 4 Neon
LMArena Text: non-English15241390Gemini 4 Neon
LMArena Text: Chinese1598Not measuredOnly Gemini 4 Neon measured
LMArena Text: French1563Not measuredOnly Gemini 4 Neon measured
LMArena Text: Spanish1505Not measuredOnly Gemini 4 Neon measured
LMArena Text: Russian15541377Gemini 4 Neon
LMArena Text: multi-turn15251406Gemini 4 Neon
LMArena Text: longer queries15331401Gemini 4 Neon
LMArena Text: writing and language15171380Gemini 4 Neon
LMArena Text: entertainment and media14951355Gemini 4 Neon
Neither model has a result on the other 29 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 4 Neon and Inkling 2 Small
BenchmarkGemini 4 NeonInkling 2 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 Neon$0.011Fits in one request
Inkling 2 Small$0.0008Fits in one request

Inkling 2 Small costs less

Repository review

50K input + 3K output tokens

Gemini 4 Neon$0.195Fits in one request
Inkling 2 Small$0.018Fits in one request

Inkling 2 Small costs less

Agent loop

200K input + 10K output tokens

Gemini 4 Neon$0.750Fits in one request
Inkling 2 Small$0.070Fits in one request

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.

Specifications of Gemini 4 Neon and Inkling 2 Small
SpecificationGoogle Gemini 4 NeonThinking Machines Inkling 2 Small
MakerGoogleThinking Machines
WeightsNo public weights listedOpen weights
ReleasedNovember 18, 2026November 9, 2026
Context window1.05M1.05M
Input $/Mtok$3.00$0.30
Output $/Mtok$15.00$1.00
Coding index86.948.8
Independent measurements behind it43
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 4 Neon better than Inkling 2 Small for coding?

Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 48.8 for Inkling 2 Small. Pick Inkling 2 Small for output tokens 15.0x cheaper ($1.00 vs $15.00 per million) and work measured by Harvey's Legal Agent Benchmark (Vals AI), where it leads.

Which is cheaper, Gemini 4 Neon or Inkling 2 Small?

Gemini 4 Neon costs $3.00 per million input tokens and $15.00 per million output tokens; Inkling 2 Small costs $0.30 and $1.00. Cost per solved task is known for neither of them.

Which has the bigger context window, Gemini 4 Neon or Inkling 2 Small?

Neither: both take 1.05M tokens.

How many benchmarks is the Gemini 4 Neon vs Inkling 2 Small comparison based on?

38 benchmarks have a result for both models as of November 22, 2026. Gemini 4 Neon'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.

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-neon-vs-thinkingmachines-inkling-2-small" title="Gemini 4 Neon vs Inkling 2 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-neon-vs-thinkingmachines-inkling-2-small">LLMBenchmark.ai, Gemini 4 Neon vs Inkling 2 Small</a></p>