Enter opens the highlighted result. Esc closes.

Gemini 4 Neon vs Step 5 Preview

Gemini 4 Neon leads Step 5 Preview on coding, 86.9 to 57.3 on the Vibe Coding Index, and Step 5 Preview costs 5.6x less per output token ($2.70 vs $15.00 per million). Updated November 22, 2026; 51 benchmarks have a result for both.

Gemini 4 NeonGoogle
Coding
86.9
Overall
89.1
Output $/M
$15.00
51
Step 5 PreviewStepFun
Coding
57.3
Overall
55.1
Output $/M
$2.70
0
vs boards won of 51 shared

Gemini 4 Neon or Step 5 Preview: which to pick

Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 57.3 for Step 5 Preview.

Pick Step 5 Preview for output tokens 5.6x cheaper ($2.70 vs $15.00 per million).

Shared results
51
Gemini 4 Neon only
18
Step 5 Preview only
25
Like-for-like categories
8 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, Step 5 Preview 57.3 from 4 benchmarks.
Widest category
Agentic work, 34.3 points apart on average over 9 shared benchmarks.
Widest single gap
LiveBench browser: Gemini 4 Neon 95.6% against Step 5 Preview 24.8%, 70.8 points apart on a 0 to 100 scale.
Output price
Gemini 4 Neon $15.00 against Step 5 Preview $2.70 per million output tokens, 5.6x apart.
Context window
Gemini 4 Neon takes 1.05M tokens, Step 5 Preview 1M.

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 NeonBarsStep 5 PreviewTies
Coding110
Intelligence10
Agentic work90
Reasoning70
Math20
Knowledge50
Web design10
General150
All shared boards510
CodingLike-for-likeSix of these boards feed the Vibe Coding Index110
Coding benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
LiveBench coding89.7%67.6%Gemini 4 Neon
LMArena WebDev17651570Gemini 4 Neon
Terminal-Bench 4 (Vals AI)61.6%31.8%Gemini 4 Neon
Vibe Code Bench (Vals AI)93.7%70.4%Gemini 4 Neon
Code migration (Vals AI)Not measured39.0%Only Step 5 Preview measured
IOI (Vals AI)Not measured38.1%Only Step 5 Preview measured
SWE-bench Multimodal (Vals AI)Not measured40.8%Only Step 5 Preview measured
SciCodeNot measured58.9%Only Step 5 Preview measured
Terminal-Bench 4.0 (Artificial Analysis run)63.6%33.3%Gemini 4 Neon
FrontierCode 2.044.4%12.0%Gemini 4 Neon
LMArena Text: coding15551481Gemini 4 Neon
LMArena Text: software and IT15421472Gemini 4 Neon
LMArena WebDev: React17871588Gemini 4 Neon
LMArena WebDev: HTML18031542Gemini 4 Neon
LMArena WebDev: from an imageNot measured1550Only Step 5 Preview measured
GameBench 298.053.2Gemini 4 Neon
CursorBench 4.065.4%Not measuredOnly Gemini 4 Neon measured
Neither model has a result on the other 35 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
Intelligence benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
AA Intelligence Index61.428.5Gemini 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 index90
Agentic work benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
LiveBench research94.5%75.5%Gemini 4 Neon
LiveBench browser95.6%24.8%Gemini 4 Neon
LiveBench long agentic82.6%19.4%Gemini 4 Neon
SREBench (Vals AI)Not measured0.4%Only Step 5 Preview measured
Finance Agent (Vals AI)Not measured50.7%Only Step 5 Preview measured
Tax Agent Bench (Vals AI)54.9%27.3%Gemini 4 Neon
Legal Research Bench (Vals AI)56.3%40.4%Gemini 4 Neon
Harvey's Legal Agent Benchmark (Vals AI)19.6%12.9%Gemini 4 Neon
Excel Modeling Benchmark (Vals AI)Not measured57.1%Only Step 5 Preview measured
BioMysteryBench (Vals AI)80.0%67.4%Gemini 4 Neon
Terminal-Bench-Science (Vals AI)62.9%2.9%Gemini 4 Neon
Vals Science Index62.6%29.6%Gemini 4 Neon
APEX-Agents82.9%Not measuredOnly Gemini 4 Neon measured
Vending-Bench 2$16,843Not measuredOnly Gemini 4 Neon measured
Neither model has a result on the other 30 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index70
Reasoning benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
LiveBench reasoning89.6%83.8%Gemini 4 Neon
MysteryMechanism (Vals AI)45.0%18.5%Gemini 4 Neon
Humanity's Last Exam (Artificial Analysis run)Not measured46.5%Only Step 5 Preview measured
HLE-Diamond (Artificial Analysis run)71.3%24.3%Gemini 4 Neon
CritPt v2, hard subset (Artificial Analysis run)61.5%32.5%Gemini 4 Neon
GPQA Diamond (Epoch AI run)96.3%Not measuredOnly Gemini 4 Neon 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 prompts15471463Gemini 4 Neon
LMArena Text: expert15601479Gemini 4 Neon
LMArena Text: hard prompts in English15241464Gemini 4 Neon
LMArena Vision: diagramsNot measured1301Only Step 5 Preview measured
LMArena Vision: homeworkNot measured1275Only Step 5 Preview measured
Neither model has a result on the other 11 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index20
Math benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
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
ProofBenchNot measured42.0%Only Step 5 Preview measured
LMArena Text: math15481466Gemini 4 Neon
LMArena Text: mathematical work15661466Gemini 4 Neon
Neither model has a result on the other 8 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index50
Knowledge benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
AA-Omniscience54.816.4Gemini 4 Neon
SimpleQA Verified75.2%Not measuredOnly Gemini 4 Neon measured
LMArena Text: science15491463Gemini 4 Neon
LMArena Text: medicine and healthcare15341443Gemini 4 Neon
LMArena Text: legal and government15181439Gemini 4 Neon
LMArena Text: business and finance14931445Gemini 4 Neon
Neither model has a result on the other 15 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index10
Web design benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
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%70.8%Gemini 4 Neon
Neither model has a result on the other 3 boards here
GeneralLike-for-likePublic boards, shown as published, kept out of the index150
General benchmarks for Gemini 4 Neon and Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
Vals Index70.9%54.8%Gemini 4 Neon
LiveBench data analysis85.3%63.5%Gemini 4 Neon
Public Benefits Bench (Vals AI)Not measured58.1%Only Step 5 Preview measured
Long context reasoning (AA-LCR)Not measured88.3%Only Step 5 Preview measured
GDPval-AANot measured1591Only Step 5 Preview measured
Furniture assembly89.2%Not measuredOnly Gemini 4 Neon measured
LMArena Text15241444Gemini 4 Neon
LMArena Text: instruction following15351446Gemini 4 Neon
LMArena Text: creative writing15191416Gemini 4 Neon
LMArena VisionNot measured1265Only Step 5 Preview measured
LMArena Text: English15111456Gemini 4 Neon
LMArena Text: non-English15241430Gemini 4 Neon
LMArena Text: Chinese15981495Gemini 4 Neon
LMArena Text: GermanNot measured1453Only Step 5 Preview measured
LMArena Text: JapaneseNot measured1398Only Step 5 Preview measured
LMArena Text: KoreanNot measured1402Only Step 5 Preview measured
LMArena Text: French15631444Gemini 4 Neon
LMArena Text: Spanish15051445Gemini 4 Neon
LMArena Text: Russian15541432Gemini 4 Neon
LMArena Text: PolishNot measured1406Only Step 5 Preview measured
LMArena Text: multi-turn15251439Gemini 4 Neon
LMArena Text: longer queries15331454Gemini 4 Neon
LMArena Text: writing and language15171431Gemini 4 Neon
LMArena Text: entertainment and media14951414Gemini 4 Neon
LMArena Vision: EnglishNot measured1275Only Step 5 Preview measured
LMArena Vision: ChineseNot measured1318Only Step 5 Preview measured
LMArena Vision: reading textNot measured1279Only Step 5 Preview measured
LMArena Vision: humorNot measured1215Only Step 5 Preview measured
LMArena Vision: creative writingNot measured1250Only Step 5 Preview measured
Neither model has a result on the other 20 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 Step 5 Preview
BenchmarkGemini 4 NeonStep 5 PreviewLeader
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
Step 5 Preview$0.0024Fits in one request

Step 5 Preview costs less

Repository review

50K input + 3K output tokens

Gemini 4 Neon$0.195Fits in one request
Step 5 Preview$0.058Fits in one request

Step 5 Preview costs less

Agent loop

200K input + 10K output tokens

Gemini 4 Neon$0.750Fits in one request
Step 5 Preview$0.227Fits in one request

Step 5 Preview 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 Step 5 Preview
SpecificationGoogle Gemini 4 NeonStepFun Step 5 Preview
MakerGoogleStepFun
WeightsNo public weights listedOpen weights
ReleasedNovember 18, 2026October 8, 2026
Context window1.05M1M
Input $/Mtok$3.00$1.00
Output $/Mtok$15.00$2.70
Coding index86.957.3
Independent measurements behind it44
Cost per solved taskNot computableNot computable
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 4 Neon better than Step 5 Preview for coding?

Pick Gemini 4 Neon for coding: it scores 86.9 on the Vibe Coding Index against 57.3 for Step 5 Preview. Pick Step 5 Preview for output tokens 5.6x cheaper ($2.70 vs $15.00 per million).

Which is cheaper, Gemini 4 Neon or Step 5 Preview?

Gemini 4 Neon costs $3.00 per million input tokens and $15.00 per million output tokens; Step 5 Preview costs $1.00 and $2.70. Cost per solved task is known for neither of them.

Which has the bigger context window, Gemini 4 Neon or Step 5 Preview?

Gemini 4 Neon, with 1.05M tokens against 1M.

How many benchmarks is the Gemini 4 Neon vs Step 5 Preview comparison based on?

51 benchmarks have a result for both models as of November 22, 2026. Gemini 4 Neon's coding index rests on 4 independent benchmarks and Step 5 Preview'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-neon-vs-stepfun-step-5-preview" title="Gemini 4 Neon vs Step 5 Preview, 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-stepfun-step-5-preview">LLMBenchmark.ai, Gemini 4 Neon vs Step 5 Preview</a></p>