Enter opens the highlighted result. Esc closes.

Gemini 3.1 Pro Preview vs Kimi K3

Kimi K3 leads Gemini 3.1 Pro Preview on coding, 62.3 to 32.3 on the Vibe Coding Index, and Gemini 3.1 Pro Preview costs 1.3x less per output token ($12.00 vs $15.00 per million). Updated October 22, 2026; 100 benchmarks have a result for both.

Replace Gemini 3.1 Pro Preview Replace Kimi K3
Gemini 3.1 Pro PreviewGoogle
Coding
32.3
Overall
54.0
Output $/M
$12.00
30
Kimi K3Moonshot AI
Coding
62.3
Overall
66.5
Output $/M
$15.00
69
vs boards won of 100 shared, 1 tied

Gemini 3.1 Pro Preview or Kimi K3: which to pick

Pick Kimi K3 for coding: it scores 62.3 on the Vibe Coding Index against 32.3 for Gemini 3.1 Pro Preview.

Pick Gemini 3.1 Pro Preview for output tokens 1.3x cheaper ($12.00 vs $15.00 per million) and work measured by LiveCodeBench (Vals AI) or Code migration (Vals AI), where it leads.

Model AGemini 3.1 Pro PreviewGoogle32.3/1004 independent benchmarks$0.99 per solved taskModel BKimi K3Moonshot AI62.3/1006 independent benchmarks$0.81 per solved task
Shared results
100
Gemini 3.1 Pro Preview only
33
Kimi K3 only
17
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 3.1 Pro Preview 32.3 from 4 benchmarks, Kimi K3 62.3 from 6 benchmarks.
Widest category
Agentic work, 18.6 points apart on average over 18 shared benchmarks.
Widest single gap
ProofBench: Gemini 3.1 Pro Preview 26.0% against Kimi K3 87.0%, 61.0 points apart on a 0 to 100 scale.
Output price
Gemini 3.1 Pro Preview $12.00 against Kimi K3 $15.00 per million output tokens, 1.3x 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 3.1 Pro PreviewBarsKimi K3Ties
Coding316
Intelligence01
Agentic work1161 tie
Reasoning103
Math06
Knowledge56
Web design07
General1114
All shared boards30691 tie
CodingLike-for-likeSix of these boards feed the Vibe Coding Index316
Coding benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
LiveBench codingNot measured59.3%Only Kimi K3 measured
LMArena WebDev14471655Kimi K3
AA Coding Index68.876.2Kimi K3
LiveCodeBench (Vals AI)88.5%87.2%Gemini 3.1 Pro Preview
SWE-bench Verified (Vals AI)78.8%93.4%Kimi K3
Terminal-Bench 4 (Vals AI)2.5%17.2%Kimi K3
Vibe Code Bench (Vals AI)32.0%85.0%Kimi K3
Code migration (Vals AI)17.3%16.1%Gemini 3.1 Pro Preview
ProgramBench (Vals AI)0.0%2.0%Kimi K3
IOI (Vals AI)51.8%48.9%Gemini 3.1 Pro Preview
Terminal-Bench 2.1 (Vals AI)70.8%80.9%Kimi K3
SWE-bench Multimodal (Vals AI)Not measured44.2%Only Kimi K3 measured
Vibe Code Bench 1-100 (Vals AI)6.7%18.2%Kimi K3
CyberBench (Vals AI)Not measured75.2%Only Kimi K3 measured
SciCode58.7%59.5%Kimi K3
Terminal-Bench Hard (Artificial Analysis run)53.8%Not measuredOnly Gemini 3.1 Pro Preview measured
Terminal-Bench80.2%Not measuredOnly Gemini 3.1 Pro Preview measured
FrontierCodeNot measured44.2%Only Kimi K3 measured
FrontierSWENot measured25.9%Only Kimi K3 measured
GSO-Bench22.6%Not measuredOnly Gemini 3.1 Pro Preview measured
WeirdML72.1%82.6%Kimi K3
MirrorCode8.9%Not measuredOnly Gemini 3.1 Pro Preview measured
ALE-Bench1160.61524.5Kimi K3
LMArena Text: coding14841508Kimi K3
LMArena Text: software and IT14871499Kimi K3
LMArena WebDev: React14381670Kimi K3
LMArena WebDev: HTML14901644Kimi K3
LMArena WebDev: from an image14811579Kimi K3
SciCode14.3%Not measuredOnly Gemini 3.1 Pro Preview measured
Neither model has a result on the other 10 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
AA Intelligence Index29.743.6Kimi K3
Epoch Capabilities IndexNot measured157.9Only Kimi K3 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index116
Agentic work benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
AA Agentic Index8.250.0Kimi K3
Design Arena: agents building web apps11251295Kimi K3
Design Arena: agents building full-stack apps10681315Kimi K3
Design Arena: agents building mobile apps10941269Kimi K3
Design Arena: agents building Android apps10491253Kimi K3
Design Arena: agents building Godot games12361223Gemini 3.1 Pro Preview
Design Arena: agents building games11131252Kimi K3
Design Arena: HTML slides11031206Kimi K3
Design Arena: PowerPoint slides10411247Kimi K3
Design Arena: spreadsheetsNot measured1070Only Kimi K3 measured
LiveBench researchNot measured79.0%Only Kimi K3 measured
LiveBench browserNot measured16.0%Only Kimi K3 measured
LiveBench long agenticNot measured42.8%Only Kimi K3 measured
SREBench (Vals AI)0.0%Not measuredOnly Gemini 3.1 Pro Preview measured
Finance Agent (Vals AI)43.0%53.1%Kimi K3
Tax Agent Bench (Vals AI)9.3%33.4%Kimi K3
Legal Research Bench (Vals AI)20.7%46.1%Kimi K3
Harvey's Legal Agent Benchmark (Vals AI)Not measured12.9%Only Kimi K3 measured
Excel Modeling Benchmark (Vals AI)52.6%66.7%Kimi K3
BioMysteryBench (Vals AI)60.0%72.6%Kimi K3
Terminal-Bench-Science (Vals AI)1.4%1.4%Tie
Vals Science Index27.4%Not measuredOnly Gemini 3.1 Pro Preview measured
Time Horizon Index (Vals AI)Not measured10.5%Only Kimi K3 measured
τ²-Bench Telecom95.6%Not measuredOnly Gemini 3.1 Pro Preview measured
APEX-Agents35.1%47.8%Kimi K3
METR time horizons77.0%Not measuredOnly Gemini 3.1 Pro Preview measured
BALROG57.0%Not measuredOnly Gemini 3.1 Pro Preview measured
DeepResearch Bench47.8%Not measuredOnly Gemini 3.1 Pro Preview measured
PostTrainBench22.0%32.0%Kimi K3
Vending-Bench 2$911$5,165Kimi K3
Neither model has a result on the other 11 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index103
Reasoning benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
GPQA Diamond95.5%92.9%Gemini 3.1 Pro Preview
LiveBench reasoningNot measured90.7%Only Kimi K3 measured
MysteryMechanism (Vals AI)20.7%Not measuredOnly Gemini 3.1 Pro Preview measured
MMMU Pro (Vals AI)88.2%88.2%Gemini 3.1 Pro Preview
GPQA Diamond (Artificial Analysis run)94.1%93.5%Gemini 3.1 Pro Preview
Humanity's Last Exam (Artificial Analysis run)47.0%46.9%Gemini 3.1 Pro Preview
GPQA Diamond (Epoch AI run)94.4%93.1%Gemini 3.1 Pro Preview
ARC-AGI98.0%94.5%Gemini 3.1 Pro Preview
ARC-AGI-277.1%60.4%Gemini 3.1 Pro Preview
SimpleBench79.6%60.7%Gemini 3.1 Pro Preview
Chess puzzles51.0%37.0%Gemini 3.1 Pro Preview
Mystery game puzzles34.0%26.0%Gemini 3.1 Pro Preview
EBR-bench14.3%Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Text: hard prompts14851495Kimi K3
LMArena Text: expert14831520Kimi K3
LMArena Text: hard prompts in English14801496Kimi K3
LMArena Vision: diagrams1311Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Vision: homework1316Not measuredOnly Gemini 3.1 Pro Preview measured
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index06
Math benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
AIME98.1%Not measuredOnly Gemini 3.1 Pro Preview measured
OTIS Mock AIME95.6%97.2%Kimi K3
FrontierMath, tiers 1 to 359.6%72.2%Kimi K3
FrontierMath, tier 426.8%39.0%Kimi K3
ProofBench26.0%87.0%Kimi K3
LMArena Text: math14861494Kimi K3
LMArena Text: mathematical work14781505Kimi K3
Neither model has a result on the other 6 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index56
Knowledge benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
MMLU Pro91.0%88.0%Gemini 3.1 Pro Preview
LegalBench (Vals AI)87.4%86.2%Gemini 3.1 Pro Preview
CaseLaw (Vals AI)64.8%Not measuredOnly Gemini 3.1 Pro Preview measured
CorpFin (Vals AI)64.5%71.6%Kimi K3
TaxEval (Vals AI)72.9%75.7%Kimi K3
MortgageTax (Vals AI)69.4%66.3%Gemini 3.1 Pro Preview
MedQA (Vals AI)96.4%Not measuredOnly Gemini 3.1 Pro Preview measured
MedCode (Vals AI)59.1%49.4%Gemini 3.1 Pro Preview
SimpleQA Verified73.5%50.6%Gemini 3.1 Pro Preview
LMArena Text: science14971508Kimi K3
LMArena Text: medicine and healthcare14861489Kimi K3
LMArena Text: legal and government14871503Kimi K3
LMArena Text: business and finance14591470Kimi K3
FACTS Benchmark Suite67.4%Not measuredOnly Gemini 3.1 Pro Preview measured
Neither model has a result on the other 4 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index07
Web design benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
Design Arena: websites12331334Kimi K3
Design Arena: UI components12241341Kimi K3
Design Arena: game dev12021358Kimi K3
Design Arena: data viz12091351Kimi K3
Design Arena: 3D12341387Kimi K3
Design Arena: all code categories12311360Kimi K3
Design Arena: SVG12851302Kimi K3
Design Arena: ASCII art1270Not measuredOnly Gemini 3.1 Pro Preview measured
LiveBench designNot measured77.2%Only Kimi K3 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index1114
General benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
Vals IndexNot measured53.0%Only Kimi K3 measured
LiveBench data analysisNot measured78.7%Only Kimi K3 measured
Public Benefits Bench (Vals AI)53.8%68.2%Kimi K3
SAGE (Vals AI)48.7%52.8%Kimi K3
MedScribe (Vals AI)76.1%88.0%Kimi K3
Vals Multimodal Index56.1%73.4%Kimi K3
IFBench77.1%Not measuredOnly Gemini 3.1 Pro Preview measured
Long context reasoning (AA-LCR)82.0%88.7%Kimi K3
DTBench97.1%91.2%Gemini 3.1 Pro Preview
LMCA53.8%52.7%Gemini 3.1 Pro Preview
CL-bench20.8%Not measuredOnly Gemini 3.1 Pro Preview measured
Surface Evolver BenchNot measured95.0%Only Kimi K3 measured
Furniture assembly26.7%34.2%Kimi K3
ExploitBench26.1%Not measuredOnly Gemini 3.1 Pro Preview measured
CL-bench Life16.9%Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Text14801476Gemini 3.1 Pro Preview
LMArena Text: instruction following14661485Kimi K3
LMArena Text: creative writing14821457Gemini 3.1 Pro Preview
LMArena Vision1296Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Text: English14781481Kimi K3
LMArena Text: non-English14761466Gemini 3.1 Pro Preview
LMArena Text: Chinese15271529Kimi K3
LMArena Text: German14901490Gemini 3.1 Pro Preview
LMArena Text: Japanese14911494Kimi K3
LMArena Text: Korean14551460Kimi K3
LMArena Text: French14871492Kimi K3
LMArena Text: Spanish14811472Gemini 3.1 Pro Preview
LMArena Text: Russian14971485Gemini 3.1 Pro Preview
LMArena Text: Polish14981458Gemini 3.1 Pro Preview
LMArena Text: multi-turn14871489Kimi K3
LMArena Text: longer queries14821494Kimi K3
LMArena Text: writing and language14811471Gemini 3.1 Pro Preview
LMArena Text: entertainment and media14651453Gemini 3.1 Pro Preview
LMArena Vision: English1287Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Vision: Chinese1364Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Vision: reading text1304Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Vision: humor1322Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Vision: captioning1345Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Vision: recognizing things1330Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Vision: creative writing1324Not measuredOnly Gemini 3.1 Pro Preview measured
LMArena Document1444Not measuredOnly Gemini 3.1 Pro Preview measured
Neither model has a result on the other 5 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 3.1 Pro Preview and Kimi K3
BenchmarkGemini 3.1 Pro PreviewKimi K3Leader
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 3.1 Pro Preview$0.0080Fits in one request
Kimi K3$0.011Fits in one request

Gemini 3.1 Pro Preview costs less

Repository review

50K input + 3K output tokens

Gemini 3.1 Pro Preview$0.136Fits in one request
Kimi K3$0.195Fits in one request

Gemini 3.1 Pro Preview costs less

Agent loop

200K input + 10K output tokens

Gemini 3.1 Pro Preview$0.520Fits in one request
Kimi K3$0.750Fits in one request

Gemini 3.1 Pro 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 3.1 Pro Preview and Kimi K3
SpecificationGoogle Gemini 3.1 Pro PreviewMoonshotAI Kimi K3
MakerGoogleMoonshot AI
WeightsNo public weights listedOpen weights
ReleasedFebruary 19, 2026July 16, 2026
Context window1.05M1.05M
Input $/Mtok$2.00$3.00
Output $/Mtok$12.00$15.00
Coding index32.362.3
Independent measurements behind it46
Cost per solved task$0.99$0.81
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 3.1 Pro Preview better than Kimi K3 for coding?

Pick Kimi K3 for coding: it scores 62.3 on the Vibe Coding Index against 32.3 for Gemini 3.1 Pro Preview. Pick Gemini 3.1 Pro Preview for output tokens 1.3x cheaper ($12.00 vs $15.00 per million) and work measured by LiveCodeBench (Vals AI) or Code migration (Vals AI), where it leads.

Which is cheaper, Gemini 3.1 Pro Preview or Kimi K3?

Gemini 3.1 Pro Preview costs $2.00 per million input tokens and $12.00 per million output tokens; Kimi K3 costs $3.00 and $15.00. Per solved SWE-bench Verified task, Gemini 3.1 Pro Preview costs $0.99 and Kimi K3 $0.81.

Which has the bigger context window, Gemini 3.1 Pro Preview or Kimi K3?

Neither: both take 1.05M tokens.

How many benchmarks is the Gemini 3.1 Pro Preview vs Kimi K3 comparison based on?

100 benchmarks have a result for both models as of October 22, 2026. Gemini 3.1 Pro Preview's coding index rests on 4 independent benchmarks and Kimi K3's on 6. 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-3-1-pro-preview-vs-kimi-k3" title="Gemini 3.1 Pro Preview vs Kimi K3, 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-3-1-pro-preview-vs-kimi-k3">LLMBenchmark.ai, Gemini 3.1 Pro Preview vs Kimi K3</a></p>