Enter opens the highlighted result. Esc closes.

Gemini 3.8 Flash vs Kimi K2.6

Updated October 21, 2026. 97 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Gemini 3.8 Flash Replace Kimi K2.6
Gemini 3.8 FlashGoogle
Coding
70.5
Overall
40.9
Output $/M
$4.125
89
Kimi K2.6Moonshot AI
Coding
44.2
Overall
27.0
Output $/M
$3.41
8
vs boards won of 97 shared

Decision reading

Pick Gemini 3.8 Flash for coding: it scores 70.5 on the Vibe Coding Index against 44.2 for Kimi K2.6.

Pick Kimi K2.6 when output tokens cost 1.2x less, or when a solved task costs $0.35 against $2.74.

Model AGemini 3.8 FlashGoogle70.5/1002 independent benchmarks$2.74 per solved taskModel BKimi K2.6Moonshot AI44.2/1001 independent benchmark$0.35 per solved task
Shared results
97
Gemini 3.8 Flash only
29
Kimi K2.6 only
12
Like-for-like categories
8 of 9

Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.

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 3.8 Flash 70.5 from 2 benchmarks, Kimi K2.6 44.2 from 1 benchmark.
Widest category
Agentic work, 14.6 points apart on average over 13 shared benchmarks.
Widest single gap
Vibe Code Bench (Vals AI): Gemini 3.8 Flash 78.7% against Kimi K2.6 37.9%, 40.8 points apart on a 0 to 100 scale.
Output price
Gemini 3.8 Flash $4.125 against Kimi K2.6 $3.41 per million output tokens, 1.2x apart.
Context window
Gemini 3.8 Flash takes 1.05M tokens, Kimi K2.6 262K.

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 3.8 FlashBarsKimi K2.6Ties
Coding160
Intelligence20
Agentic work94
Reasoning120
Math51
Knowledge82
Web design80
General291
All shared boards898
CodingLike-for-likeFour of these boards feed the Vibe Coding Index160
Coding benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
LiveBench coding82.2%Not measuredOnly Gemini 3.8 Flash measured
LMArena WebDev15851509Gemini 3.8 Flash
AA Coding Index76.361.8Gemini 3.8 Flash
LiveCodeBench (Vals AI)89.5%86.8%Gemini 3.8 Flash
SWE-bench Verified (Vals AI)80.0%76.2%Gemini 3.8 Flash
Terminal-Bench 4 (Vals AI)19.7%Not measuredOnly Gemini 3.8 Flash measured
Vibe Code Bench (Vals AI)78.7%37.9%Gemini 3.8 Flash
Code migration (Vals AI)36.5%27.8%Gemini 3.8 Flash
ProgramBench (Vals AI)1.0%0.0%Gemini 3.8 Flash
IOI (Vals AI)56.9%Not measuredOnly Gemini 3.8 Flash measured
Terminal-Bench 2.1 (Vals AI)81.3%53.6%Gemini 3.8 Flash
SWE-bench Multimodal (Vals AI)40.8%Not measuredOnly Gemini 3.8 Flash measured
Vibe Code Bench 1-100 (Vals AI)18.8%Not measuredOnly Gemini 3.8 Flash measured
SkillsBench (Vals AI)58.0%Not measuredOnly Gemini 3.8 Flash measured
CyberBench (Vals AI)43.8%Not measuredOnly Gemini 3.8 Flash measured
SciCode56.6%51.5%Gemini 3.8 Flash
Terminal-Bench Hard (Artificial Analysis run)Not measured43.9%Only Kimi K2.6 measured
FrontierCode41.2%Not measuredOnly Gemini 3.8 Flash measured
FrontierSWE19.6%Not measuredOnly Gemini 3.8 Flash measured
WeirdML84.8%55.9%Gemini 3.8 Flash
ALE-Bench1269.71092.7Gemini 3.8 Flash
LMArena Text: coding15081489Gemini 3.8 Flash
LMArena Text: software and IT15071484Gemini 3.8 Flash
LMArena WebDev: React15771505Gemini 3.8 Flash
LMArena WebDev: HTML16161485Gemini 3.8 Flash
LMArena WebDev: from an image15931521Gemini 3.8 Flash
Neither model has a result on the other 12 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
Intelligence benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
AA Intelligence Index40.927.0Gemini 3.8 Flash
Epoch Capabilities Index157.4150.9Gemini 3.8 Flash
Agentic workLike-for-likePublic boards, shown as published, kept out of the index94
Agentic work benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
AA Agentic Index40.220.5Gemini 3.8 Flash
Design Arena: agents building web apps12381268Kimi K2.6
Design Arena: agents building full-stack apps12441155Gemini 3.8 Flash
Design Arena: agents building mobile apps12361156Gemini 3.8 Flash
Design Arena: agents building Android apps10781166Kimi K2.6
Design Arena: agents building Godot games12731120Gemini 3.8 Flash
Design Arena: agents building gamesNot measured1141Only Kimi K2.6 measured
Design Arena: HTML slidesNot measured1184Only Kimi K2.6 measured
Design Arena: PowerPoint slides11781128Gemini 3.8 Flash
Design Arena: spreadsheets10521276Kimi K2.6
LiveBench research79.1%Not measuredOnly Gemini 3.8 Flash measured
LiveBench browser64.0%Not measuredOnly Gemini 3.8 Flash measured
LiveBench long agentic30.1%Not measuredOnly Gemini 3.8 Flash measured
SREBench (Vals AI)11.1%Not measuredOnly Gemini 3.8 Flash measured
Finance Agent (Vals AI)61.4%44.9%Gemini 3.8 Flash
Tax Agent Bench (Vals AI)32.1%12.4%Gemini 3.8 Flash
Legal Research Bench (Vals AI)38.9%15.9%Gemini 3.8 Flash
Harvey's Legal Agent Benchmark (Vals AI)12.1%Not measuredOnly Gemini 3.8 Flash measured
Excel Modeling Benchmark (Vals AI)72.2%57.9%Gemini 3.8 Flash
BioMysteryBench (Vals AI)62.2%Not measuredOnly Gemini 3.8 Flash measured
Terminal-Bench-Science (Vals AI)8.6%Not measuredOnly Gemini 3.8 Flash measured
Vals Science Index35.8%Not measuredOnly Gemini 3.8 Flash measured
Time Horizon Index (Vals AI)18.8%Not measuredOnly Gemini 3.8 Flash measured
CUA-bench (Vals AI)4.2%Not measuredOnly Gemini 3.8 Flash measured
τ²-Bench TelecomNot measured95.9%Only Kimi K2.6 measured
APEX-Agents64.3%Not measuredOnly Gemini 3.8 Flash measured
Remote Labor Index5.8%Not measuredOnly Gemini 3.8 Flash measured
OSWorld 2.0Not measured4.6%Only Kimi K2.6 measured
Vending-Bench 2$5,094$6,205Kimi K2.6
Neither model has a result on the other 10 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index120
Reasoning benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
GPQA Diamond94.4%89.1%Gemini 3.8 Flash
LiveBench reasoning89.3%Not measuredOnly Gemini 3.8 Flash measured
MysteryMechanism (Vals AI)36.5%Not measuredOnly Gemini 3.8 Flash measured
MMMU Pro (Vals AI)89.1%86.3%Gemini 3.8 Flash
GPQA Diamond (Artificial Analysis run)95.3%91.1%Gemini 3.8 Flash
Humanity's Last Exam (Artificial Analysis run)47.8%37.5%Gemini 3.8 Flash
GPQA Diamond (Epoch AI run)95.4%90.8%Gemini 3.8 Flash
ARC-AGI98.5%Not measuredOnly Gemini 3.8 Flash measured
ARC-AGI-289.2%Not measuredOnly Gemini 3.8 Flash measured
Chess puzzles58.0%20.0%Gemini 3.8 Flash
Mystery game puzzles47.0%18.0%Gemini 3.8 Flash
EBR-benchNot measured2.4%Only Kimi K2.6 measured
LMArena Text: hard prompts15091470Gemini 3.8 Flash
LMArena Text: expert15261491Gemini 3.8 Flash
LMArena Text: hard prompts in English15041471Gemini 3.8 Flash
LMArena Vision: diagrams13451298Gemini 3.8 Flash
LMArena Vision: homework13511310Gemini 3.8 Flash
Neither model has a result on the other 4 boards here
MathLike-for-likePublic boards, shown as published, kept out of the index51
Math benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
OTIS Mock AIME98.9%96.1%Gemini 3.8 Flash
FrontierMath, tiers 1 to 368.4%57.2%Gemini 3.8 Flash
FrontierMath, tier 422.0%25.6%Kimi K2.6
ProofBench48.0%16.0%Gemini 3.8 Flash
LMArena Text: math15321474Gemini 3.8 Flash
LMArena Text: mathematical work15191480Gemini 3.8 Flash
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index82
Knowledge benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
MMLU Pro90.2%87.6%Gemini 3.8 Flash
LegalBench (Vals AI)87.0%84.7%Gemini 3.8 Flash
CaseLaw (Vals AI)Not measured61.2%Only Kimi K2.6 measured
CorpFin (Vals AI)Not measured66.7%Only Kimi K2.6 measured
TaxEval (Vals AI)74.5%74.7%Kimi K2.6
MortgageTax (Vals AI)65.3%65.8%Kimi K2.6
MedCode (Vals AI)48.1%40.1%Gemini 3.8 Flash
SimpleQA Verified69.7%34.9%Gemini 3.8 Flash
LMArena Text: science15231468Gemini 3.8 Flash
LMArena Text: medicine and healthcare15081457Gemini 3.8 Flash
LMArena Text: legal and government14971462Gemini 3.8 Flash
LMArena Text: business and finance14791445Gemini 3.8 Flash
Neither model has a result on the other 5 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index80
Web design benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
Design Arena: websites13011269Gemini 3.8 Flash
Design Arena: UI components13181266Gemini 3.8 Flash
Design Arena: game dev13051251Gemini 3.8 Flash
Design Arena: data viz12691249Gemini 3.8 Flash
Design Arena: 3D12951268Gemini 3.8 Flash
Design Arena: all code categories13041272Gemini 3.8 Flash
Design Arena: SVG12951181Gemini 3.8 Flash
Design Arena: ASCII art12641164Gemini 3.8 Flash
LiveBench design72.7%Not measuredOnly Gemini 3.8 Flash measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index291
General benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
Vals Index56.0%Not measuredOnly Gemini 3.8 Flash measured
LiveBench data analysis54.0%Not measuredOnly Gemini 3.8 Flash measured
Public Benefits Bench (Vals AI)65.3%56.6%Gemini 3.8 Flash
SAGE (Vals AI)35.1%50.2%Kimi K2.6
MedScribe (Vals AI)84.5%78.2%Gemini 3.8 Flash
Vals Multimodal IndexNot measured56.4%Only Kimi K2.6 measured
IFBenchNot measured76.0%Only Kimi K2.6 measured
Long context reasoning (AA-LCR)84.0%81.0%Gemini 3.8 Flash
DTBench95.7%90.9%Gemini 3.8 Flash
LMCA52.9%37.3%Gemini 3.8 Flash
Surface Evolver Bench76.9%Not measuredOnly Gemini 3.8 Flash measured
Furniture assembly31.7%21.7%Gemini 3.8 Flash
ExploitBenchNot measured18.4%Only Kimi K2.6 measured
LMArena Text14991455Gemini 3.8 Flash
LMArena Text: instruction following14921451Gemini 3.8 Flash
LMArena Text: creative writing14881434Gemini 3.8 Flash
LMArena Vision13151282Gemini 3.8 Flash
LMArena Text: English14981460Gemini 3.8 Flash
LMArena Text: non-English14911446Gemini 3.8 Flash
LMArena Text: Chinese15531520Gemini 3.8 Flash
LMArena Text: German15011451Gemini 3.8 Flash
LMArena Text: Japanese15071443Gemini 3.8 Flash
LMArena Text: Korean14641427Gemini 3.8 Flash
LMArena Text: French14961471Gemini 3.8 Flash
LMArena Text: Spanish14871464Gemini 3.8 Flash
LMArena Text: Russian15141446Gemini 3.8 Flash
LMArena Text: Polish14971438Gemini 3.8 Flash
LMArena Text: multi-turn15031453Gemini 3.8 Flash
LMArena Text: longer queries15071468Gemini 3.8 Flash
LMArena Text: writing and language14941439Gemini 3.8 Flash
LMArena Text: entertainment and media14801428Gemini 3.8 Flash
LMArena Vision: English13181287Gemini 3.8 Flash
LMArena Vision: Chinese14161346Gemini 3.8 Flash
LMArena Vision: reading text13331291Gemini 3.8 Flash
LMArena Vision: humor13211280Gemini 3.8 Flash
LMArena Vision: creative writing12991273Gemini 3.8 Flash
LMArena DocumentNot measured1451Only Kimi K2.6 measured
Neither model has a result on the other 9 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Gemini 3.8 Flash and Kimi K2.6
BenchmarkGemini 3.8 FlashKimi K2.6Leader
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 3.8 Flash$0.0029Fits in one request
Kimi K2.6$0.0024Fits in one request

Kimi K2.6 costs less

Repository review

50K input + 3K output tokens

Gemini 3.8 Flash$0.054Fits in one request
Kimi K2.6$0.043Fits in one request

Kimi K2.6 costs less

Agent loop

200K input + 10K output tokens

Gemini 3.8 Flash$0.206Fits in one request
Kimi K2.6$0.164Fits in one request

Kimi K2.6 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 3.8 Flash and Kimi K2.6
SpecificationGoogle Gemini 3.8 FlashMoonshotAI Kimi K2.6
MakerGoogleMoonshot AI
WeightsNo public weights listedOpen weights
ReleasedSeptember 2, 2026April 20, 2026
Context window1.05M262K
Input $/Mtok$0.825$0.65
Output $/Mtok$4.125$3.41
Coding index70.544.2
Independent measurements behind it21
Cost per solved task$2.74$0.35
SourceOpenRouter listingOpenRouter listing

Questions

Is Gemini 3.8 Flash better than Kimi K2.6 for coding?

Pick Gemini 3.8 Flash for coding: it scores 70.5 on the Vibe Coding Index against 44.2 for Kimi K2.6. Pick Kimi K2.6 when output tokens cost 1.2x less, or when a solved task costs $0.35 against $2.74.

Which is cheaper, Gemini 3.8 Flash or Kimi K2.6?

Gemini 3.8 Flash costs $0.825 per million input tokens and $4.125 per million output tokens; Kimi K2.6 costs $0.65 and $3.41. Per solved SWE-bench Verified task, Gemini 3.8 Flash costs $2.74 and Kimi K2.6 $0.35.

Which has the bigger context window, Gemini 3.8 Flash or Kimi K2.6?

Gemini 3.8 Flash, with 1.05M tokens against 262K.

How many benchmarks is the Gemini 3.8 Flash vs Kimi K2.6 comparison based on?

97 benchmarks have a result for both models as of October 21, 2026. Gemini 3.8 Flash's coding index rests on 2 independent benchmarks and Kimi K2.6's on 1. Where only one model has a result, the page says so instead of naming a winner.

Compare next