Enter opens the highlighted result. Esc closes.

Step 5 Preview vs Inkling 2 Mini

Step 5 Preview leads Inkling 2 Mini on coding, 57.3 to 39.5 on the Vibe Coding Index, and Inkling 2 Mini costs 8.3x less per output token ($0.325 vs $2.70 per million). Updated November 22, 2026; 20 benchmarks have a result for both.

Step 5 PreviewStepFun
Coding
57.3
Overall
55.1
Output $/M
$2.70
17
T Inkling 2 MiniThinking Machines
Coding
39.5
Overall
No data
Output $/M
$0.325
3
vs boards won of 20 shared

Step 5 Preview or Inkling 2 Mini: which to pick

Pick Step 5 Preview for coding: it scores 57.3 on the Vibe Coding Index against 39.5 for Inkling 2 Mini.

Pick Inkling 2 Mini for output tokens 8.3x cheaper ($0.325 vs $2.70 per million) and a larger context window (1.05M against 1M tokens).

Shared results
20
Step 5 Preview only
56
Inkling 2 Mini only
9
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
Step 5 Preview 57.3 from 4 benchmarks, Inkling 2 Mini 39.5 from 2 benchmarks.
Widest category
Coding, 16.9 points apart on average over 6 shared benchmarks.
Widest single gap
Terminal-Bench 4 (Vals AI): Step 5 Preview 31.8% against Inkling 2 Mini 6.6%, 33.7 points apart on a 0 to 100 scale.
Output price
Step 5 Preview $2.70 against Inkling 2 Mini $0.325 per million output tokens, 8.3x apart.
Context window
Step 5 Preview takes 1M 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
CategoryStep 5 PreviewBarsInkling 2 MiniTies
Coding42
Intelligence10
Agentic work70
Reasoning21
Knowledge10
General20
All shared boards173
CodingLike-for-likeSix of these boards feed the Vibe Coding Index42
Coding benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
LiveBench coding67.6%Not measuredOnly Step 5 Preview measured
LMArena WebDev1570Not measuredOnly Step 5 Preview measured
Terminal-Bench 4 (Vals AI)31.8%6.6%Step 5 Preview
Vibe Code Bench (Vals AI)70.4%77.9%Inkling 2 Mini
Code migration (Vals AI)39.0%Not measuredOnly Step 5 Preview measured
ProgramBench (Vals AI)Not measured0.5%Only Inkling 2 Mini measured
IOI (Vals AI)38.1%11.6%Step 5 Preview
SWE-bench Multimodal (Vals AI)40.8%37.9%Step 5 Preview
SkillsBench (Vals AI)Not measured43.2%Only Inkling 2 Mini measured
CyberBench (Vals AI)Not measured74.6%Only Inkling 2 Mini measured
SciCode58.9%Not measuredOnly Step 5 Preview measured
Terminal-Bench 4.0 (Artificial Analysis run)33.3%5.6%Step 5 Preview
FrontierCode 2.012.0%15.1%Inkling 2 Mini
LMArena Text: coding1481Not measuredOnly Step 5 Preview measured
LMArena Text: software and IT1472Not measuredOnly Step 5 Preview measured
LMArena WebDev: React1588Not measuredOnly Step 5 Preview measured
LMArena WebDev: HTML1542Not measuredOnly Step 5 Preview measured
LMArena WebDev: from an image1550Not measuredOnly Step 5 Preview measured
GameBench 253.2Not measuredOnly Step 5 Preview measured
Neither model has a result on the other 33 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
AA Intelligence Index28.520.1Step 5 Preview
Neither model has a result on the other 1 board here
Agentic workLike-for-likePublic boards, shown as published, kept out of the index70
Agentic work benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
LiveBench research75.5%Not measuredOnly Step 5 Preview measured
LiveBench browser24.8%Not measuredOnly Step 5 Preview measured
LiveBench long agentic19.4%Not measuredOnly Step 5 Preview measured
SREBench (Vals AI)0.4%0.0%Step 5 Preview
Finance Agent (Vals AI)50.7%Not measuredOnly Step 5 Preview measured
Tax Agent Bench (Vals AI)27.3%14.1%Step 5 Preview
Legal Research Bench (Vals AI)40.4%17.8%Step 5 Preview
Harvey's Legal Agent Benchmark (Vals AI)12.9%5.0%Step 5 Preview
Excel Modeling Benchmark (Vals AI)57.1%Not measuredOnly Step 5 Preview measured
BioMysteryBench (Vals AI)67.4%47.8%Step 5 Preview
Terminal-Bench-Science (Vals AI)2.9%0.0%Step 5 Preview
Vals Science Index29.6%23.6%Step 5 Preview
Neither model has a result on the other 32 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index21
Reasoning benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
LiveBench reasoning83.8%Not measuredOnly Step 5 Preview measured
MysteryMechanism (Vals AI)18.5%23.0%Inkling 2 Mini
Humanity's Last Exam (Artificial Analysis run)46.5%Not measuredOnly Step 5 Preview measured
HLE-Diamond (Artificial Analysis run)24.3%15.2%Step 5 Preview
CritPt v2, hard subset (Artificial Analysis run)32.5%17.0%Step 5 Preview
LMArena Text: hard prompts1463Not measuredOnly Step 5 Preview measured
LMArena Text: expert1479Not measuredOnly Step 5 Preview measured
LMArena Text: hard prompts in English1464Not measuredOnly Step 5 Preview measured
LMArena Vision: diagrams1301Not measuredOnly Step 5 Preview measured
LMArena Vision: homework1275Not measuredOnly Step 5 Preview measured
Neither model has a result on the other 16 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
Math benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
ProofBench42.0%Not measuredOnly Step 5 Preview measured
LMArena Text: math1466Not measuredOnly Step 5 Preview measured
LMArena Text: mathematical work1466Not measuredOnly Step 5 Preview measured
Neither model has a result on the other 11 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index10
Knowledge benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
MedCode (Vals AI)Not measured25.7%Only Inkling 2 Mini measured
AA-Omniscience16.4-8.6Step 5 Preview
LMArena Text: science1463Not measuredOnly Step 5 Preview measured
LMArena Text: medicine and healthcare1443Not measuredOnly Step 5 Preview measured
LMArena Text: legal and government1439Not measuredOnly Step 5 Preview measured
LMArena Text: business and finance1445Not measuredOnly Step 5 Preview measured
Neither model has a result on the other 15 boards here
Web designDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
Web design benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
Design Arena: UI componentsNot measured1292Only Inkling 2 Mini measured
Design Arena: game devNot measured1287Only Inkling 2 Mini measured
Design Arena: 3DNot measured1308Only Inkling 2 Mini measured
LiveBench design70.8%Not measuredOnly Step 5 Preview measured
Neither model has a result on the other 5 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index20
General benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 2 MiniLeader
Vals Index54.8%46.5%Step 5 Preview
LiveBench data analysis63.5%Not measuredOnly Step 5 Preview measured
Public Benefits Bench (Vals AI)58.1%Not measuredOnly Step 5 Preview measured
SAGE (Vals AI)Not measured85.9%Only Inkling 2 Mini measured
MedScribe (Vals AI)Not measured86.4%Only Inkling 2 Mini measured
Long context reasoning (AA-LCR)88.3%Not measuredOnly Step 5 Preview measured
GDPval-AA15911319Step 5 Preview
LMArena Text1444Not measuredOnly Step 5 Preview measured
LMArena Text: instruction following1446Not measuredOnly Step 5 Preview measured
LMArena Text: creative writing1416Not measuredOnly Step 5 Preview measured
LMArena Vision1265Not measuredOnly Step 5 Preview measured
LMArena Text: English1456Not measuredOnly Step 5 Preview measured
LMArena Text: non-English1430Not measuredOnly Step 5 Preview measured
LMArena Text: Chinese1495Not measuredOnly Step 5 Preview measured
LMArena Text: German1453Not measuredOnly Step 5 Preview measured
LMArena Text: Japanese1398Not measuredOnly Step 5 Preview measured
LMArena Text: Korean1402Not measuredOnly Step 5 Preview measured
LMArena Text: French1444Not measuredOnly Step 5 Preview measured
LMArena Text: Spanish1445Not measuredOnly Step 5 Preview measured
LMArena Text: Russian1432Not measuredOnly Step 5 Preview measured
LMArena Text: Polish1406Not measuredOnly Step 5 Preview measured
LMArena Text: multi-turn1439Not measuredOnly Step 5 Preview measured
LMArena Text: longer queries1454Not measuredOnly Step 5 Preview measured
LMArena Text: writing and language1431Not measuredOnly Step 5 Preview measured
LMArena Text: entertainment and media1414Not measuredOnly Step 5 Preview measured
LMArena Vision: English1275Not measuredOnly Step 5 Preview measured
LMArena Vision: Chinese1318Not measuredOnly Step 5 Preview measured
LMArena Vision: reading text1279Not measuredOnly Step 5 Preview measured
LMArena Vision: humor1215Not measuredOnly Step 5 Preview measured
LMArena Vision: creative writing1250Not measuredOnly Step 5 Preview measured
Neither model has a result on the other 19 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Step 5 Preview and Inkling 2 Mini
BenchmarkStep 5 PreviewInkling 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

Step 5 Preview$0.0024Fits in one request
Inkling 2 Mini$0.0002Fits in one request

Inkling 2 Mini costs less

Repository review

50K input + 3K output tokens

Step 5 Preview$0.058Fits in one request
Inkling 2 Mini$0.0049Fits in one request

Inkling 2 Mini costs less

Agent loop

200K input + 10K output tokens

Step 5 Preview$0.227Fits 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 Step 5 Preview and Inkling 2 Mini
SpecificationStepFun Step 5 PreviewThinking Machines Inkling 2 Mini
MakerStepFunThinking Machines
WeightsOpen weightsOpen weights
ReleasedOctober 8, 2026November 21, 2026
Context window1M1.05M
Input $/Mtok$1.00$0.0775
Output $/Mtok$2.70$0.325
Coding index57.339.5
Independent measurements behind it42
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Step 5 Preview better than Inkling 2 Mini for coding?

Pick Step 5 Preview for coding: it scores 57.3 on the Vibe Coding Index against 39.5 for Inkling 2 Mini. Pick Inkling 2 Mini for output tokens 8.3x cheaper ($0.325 vs $2.70 per million) and a larger context window (1.05M against 1M tokens).

Which is cheaper, Step 5 Preview or Inkling 2 Mini?

Step 5 Preview costs $1.00 per million input tokens and $2.70 per million output tokens; Inkling 2 Mini costs $0.0775 and $0.325. Cost per solved task is known for neither of them.

Which has the bigger context window, Step 5 Preview or Inkling 2 Mini?

Inkling 2 Mini, with 1.05M tokens against 1M.

How many benchmarks is the Step 5 Preview vs Inkling 2 Mini comparison based on?

20 benchmarks have a result for both models as of November 22, 2026. Step 5 Preview's coding index rests on 4 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/stepfun-step-5-preview-vs-thinkingmachines-inkling-2-mini" title="Step 5 Preview 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/stepfun-step-5-preview-vs-thinkingmachines-inkling-2-mini">LLMBenchmark.ai, Step 5 Preview vs Inkling 2 Mini</a></p>