LLMBenchmark.ai

Enter opens the highlighted result. Esc closes.

Mercury 2 vs Kimi K3

Updated October 21, 2026. 38 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.

Replace Mercury 2 Replace Kimi K3
Mercury 2Inception
Coding
0.0
Overall
10.8
Output $/M
$0.75
0
Kimi K3Moonshot AI
Coding
61.8
Overall
43.6
Output $/M
$11.936
38
vs boards won of 38 shared

Decision reading

Pick Kimi K3 for coding: it scores 61.8 on the Vibe Coding Index against 0.0 for Mercury 2.

Pick Mercury 2 when output tokens cost 15.9x less.

Model AMercury 2Inception0.0/1001 independent benchmarkNo cost per solved taskModel BKimi K3Moonshot AI61.8/1002 independent benchmarks$0.81 per solved task
Shared results
38
Mercury 2 only
4
Kimi K3 only
79
Like-for-like categories
7 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
Mercury 2 0.0 from 1 benchmark, Kimi K3 61.8 from 2 benchmarks.
Widest category
Agentic work, 48.4 points apart on average over 1 shared benchmark.
Widest single gap
LMArena WebDev: React: Mercury 2 1160 against Kimi K3 1670, 73.4 points apart on a 0 to 100 scale.
Output price
Mercury 2 $0.75 against Kimi K3 $11.936 per million output tokens, 15.9x apart.
Context window
Mercury 2 takes 128K tokens, Kimi K3 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
CategoryMercury 2BarsKimi K3Ties
Coding09
Intelligence01
Agentic work01
Reasoning05
Knowledge03
Web design07
General012
All shared boards038
CodingLike-for-likeFour of these boards feed the Vibe Coding Index09
Coding benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
LiveBench codingNot measured59.3%Only Kimi K3 measured
LMArena WebDev11711655Kimi K3
AA Coding Index31.176.2Kimi K3
LiveCodeBench (Vals AI)Not measured87.2%Only Kimi K3 measured
SWE-bench Verified (Vals AI)Not measured93.4%Only Kimi K3 measured
Terminal-Bench 4 (Vals AI)Not measured17.2%Only Kimi K3 measured
Vibe Code Bench (Vals AI)Not measured85.0%Only Kimi K3 measured
Code migration (Vals AI)Not measured16.1%Only Kimi K3 measured
ProgramBench (Vals AI)Not measured2.0%Only Kimi K3 measured
IOI (Vals AI)Not measured48.9%Only Kimi K3 measured
Terminal-Bench 2.1 (Vals AI)Not measured80.9%Only Kimi K3 measured
SWE-bench Multimodal (Vals AI)Not measured44.2%Only Kimi K3 measured
Vibe Code Bench 1-100 (Vals AI)Not measured18.2%Only Kimi K3 measured
CyberBench (Vals AI)Not measured75.2%Only Kimi K3 measured
SciCode37.7%59.5%Kimi K3
Terminal-Bench Hard (Artificial Analysis run)26.5%Not measuredOnly Mercury 2 measured
FrontierCodeNot measured44.2%Only Kimi K3 measured
FrontierSWENot measured25.9%Only Kimi K3 measured
WeirdML43.2%82.6%Kimi K3
ALE-Bench785.61524.5Kimi K3
LMArena Text: coding13901508Kimi K3
LMArena Text: software and IT13851499Kimi K3
LMArena WebDev: React11601670Kimi K3
LMArena WebDev: HTML11991644Kimi K3
LMArena WebDev: from an imageNot measured1579Only Kimi K3 measured
Neither model has a result on the other 13 boards here
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
Intelligence benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
AA Intelligence Index10.843.6Kimi K3
Epoch Capabilities IndexNot measured157.9Only Kimi K3 measured
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
Agentic work benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
AA Agentic Index1.650.0Kimi K3
Design Arena: agents building web appsNot measured1296Only Kimi K3 measured
Design Arena: agents building full-stack appsNot measured1315Only Kimi K3 measured
Design Arena: agents building mobile appsNot measured1267Only Kimi K3 measured
Design Arena: agents building Android appsNot measured1256Only Kimi K3 measured
Design Arena: agents building Godot gamesNot measured1223Only Kimi K3 measured
Design Arena: agents building gamesNot measured1252Only Kimi K3 measured
Design Arena: HTML slidesNot measured1256Only Kimi K3 measured
Design Arena: PowerPoint slidesNot measured1273Only Kimi K3 measured
Design Arena: spreadsheetsNot measured1058Only 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
Finance Agent (Vals AI)Not measured53.1%Only Kimi K3 measured
Tax Agent Bench (Vals AI)Not measured33.4%Only Kimi K3 measured
Legal Research Bench (Vals AI)Not measured46.1%Only Kimi K3 measured
Harvey's Legal Agent Benchmark (Vals AI)Not measured12.9%Only Kimi K3 measured
Excel Modeling Benchmark (Vals AI)Not measured66.7%Only Kimi K3 measured
BioMysteryBench (Vals AI)Not measured72.6%Only Kimi K3 measured
Terminal-Bench-Science (Vals AI)Not measured1.4%Only Kimi K3 measured
Time Horizon Index (Vals AI)Not measured10.5%Only Kimi K3 measured
τ²-Bench Telecom70.8%Not measuredOnly Mercury 2 measured
APEX-AgentsNot measured47.8%Only Kimi K3 measured
PostTrainBenchNot measured32.0%Only Kimi K3 measured
Vending-Bench 2Not measured$5,165Only Kimi K3 measured
Neither model has a result on the other 14 boards here
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
Reasoning benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
GPQA DiamondNot measured92.9%Only Kimi K3 measured
LiveBench reasoningNot measured90.7%Only Kimi K3 measured
MMMU Pro (Vals AI)Not measured88.2%Only Kimi K3 measured
GPQA Diamond (Artificial Analysis run)77.0%93.5%Kimi K3
Humanity's Last Exam (Artificial Analysis run)17.1%46.9%Kimi K3
GPQA Diamond (Epoch AI run)Not measured93.1%Only Kimi K3 measured
ARC-AGINot measured94.5%Only Kimi K3 measured
ARC-AGI-2Not measured60.4%Only Kimi K3 measured
SimpleBenchNot measured60.7%Only Kimi K3 measured
Chess puzzlesNot measured37.0%Only Kimi K3 measured
Mystery game puzzlesNot measured26.0%Only Kimi K3 measured
LMArena Text: hard prompts13621495Kimi K3
LMArena Text: expert13581520Kimi K3
LMArena Text: hard prompts in English13741496Kimi K3
Neither model has a result on the other 7 boards here
MathDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
Math benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
OTIS Mock AIMENot measured97.2%Only Kimi K3 measured
FrontierMath, tiers 1 to 3Not measured72.2%Only Kimi K3 measured
FrontierMath, tier 4Not measured39.0%Only Kimi K3 measured
ProofBenchNot measured87.0%Only Kimi K3 measured
LMArena Text: mathNot measured1494Only Kimi K3 measured
LMArena Text: mathematical workNot measured1505Only Kimi K3 measured
Neither model has a result on the other 7 boards here
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index03
Knowledge benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
MMLU ProNot measured88.0%Only Kimi K3 measured
LegalBench (Vals AI)Not measured86.2%Only Kimi K3 measured
CorpFin (Vals AI)Not measured71.6%Only Kimi K3 measured
TaxEval (Vals AI)Not measured75.7%Only Kimi K3 measured
MortgageTax (Vals AI)Not measured66.3%Only Kimi K3 measured
MedCode (Vals AI)Not measured49.4%Only Kimi K3 measured
SimpleQA VerifiedNot measured50.6%Only Kimi K3 measured
LMArena Text: science13431508Kimi K3
LMArena Text: medicine and healthcare13631489Kimi K3
LMArena Text: legal and governmentNot measured1503Only Kimi K3 measured
LMArena Text: business and finance13491470Kimi K3
Neither model has a result on the other 6 boards here
Web designLike-for-likePublic boards, shown as published, kept out of the index07
Web design benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
Design Arena: websites10361336Kimi K3
Design Arena: UI components9981345Kimi K3
Design Arena: game dev9851361Kimi K3
Design Arena: data viz10091350Kimi K3
Design Arena: 3D9821388Kimi K3
Design Arena: all code categories10221361Kimi K3
Design Arena: SVG9981301Kimi K3
Design Arena: ASCII art990Not measuredOnly Mercury 2 measured
LiveBench designNot measured77.2%Only Kimi K3 measured
GeneralLike-for-likePublic boards, shown as published, kept out of the index012
General benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi K3Leader
Vals IndexNot measured53.0%Only Kimi K3 measured
LiveBench data analysisNot measured78.7%Only Kimi K3 measured
Public Benefits Bench (Vals AI)Not measured68.2%Only Kimi K3 measured
SAGE (Vals AI)Not measured52.8%Only Kimi K3 measured
MedScribe (Vals AI)Not measured88.0%Only Kimi K3 measured
Vals Multimodal IndexNot measured73.4%Only Kimi K3 measured
IFBench69.8%Not measuredOnly Mercury 2 measured
Long context reasoning (AA-LCR)43.7%88.7%Kimi K3
DTBenchNot measured91.2%Only Kimi K3 measured
LMCANot measured52.7%Only Kimi K3 measured
Surface Evolver BenchNot measured95.0%Only Kimi K3 measured
Furniture assemblyNot measured34.2%Only Kimi K3 measured
LMArena Text13541476Kimi K3
LMArena Text: instruction following13281485Kimi K3
LMArena Text: creative writing12861457Kimi K3
LMArena Text: English13781481Kimi K3
LMArena Text: non-English13301466Kimi K3
LMArena Text: Chinese14141529Kimi K3
LMArena Text: GermanNot measured1490Only Kimi K3 measured
LMArena Text: JapaneseNot measured1494Only Kimi K3 measured
LMArena Text: KoreanNot measured1460Only Kimi K3 measured
LMArena Text: FrenchNot measured1492Only Kimi K3 measured
LMArena Text: SpanishNot measured1472Only Kimi K3 measured
LMArena Text: Russian13041485Kimi K3
LMArena Text: PolishNot measured1458Only Kimi K3 measured
LMArena Text: multi-turn13571489Kimi K3
LMArena Text: longer queries13291494Kimi K3
LMArena Text: writing and language13001471Kimi K3
LMArena Text: entertainment and media13141453Kimi K3
Neither model has a result on the other 17 boards here
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
Image generation benchmarks for Mercury 2 and Kimi K3
BenchmarkMercury 2Kimi 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

Mercury 2$0.0006Fits in one request
Kimi K3$0.0062Fits in one request

Mercury 2 costs less

Repository review

50K input + 3K output tokens

Mercury 2$0.015Fits in one request
Kimi K3$0.048Fits in one request

Mercury 2 costs less

Agent loop

200K input + 10K output tokens

Mercury 2$0.058Exceeds the context window
Kimi K3$0.167Fits in one request

Mercury 2 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 Mercury 2 and Kimi K3
SpecificationInception Mercury 2MoonshotAI Kimi K3
MakerInceptionMoonshot AI
WeightsNo public weights listedOpen weights
ReleasedMarch 4, 2026July 16, 2026
Context window128K1.05M
Input $/Mtok$0.25$0.239
Output $/Mtok$0.75$11.936
Coding index0.061.8
Independent measurements behind it12
Cost per solved taskNot computable$0.81
SourceOpenRouter listingOpenRouter listing

Questions

Is Mercury 2 better than Kimi K3 for coding?

Pick Kimi K3 for coding: it scores 61.8 on the Vibe Coding Index against 0.0 for Mercury 2. Pick Mercury 2 when output tokens cost 15.9x less.

Which is cheaper, Mercury 2 or Kimi K3?

Mercury 2 costs $0.25 per million input tokens and $0.75 per million output tokens; Kimi K3 costs $0.239 and $11.936. Cost per solved task is known for Kimi K3 only.

Which has the bigger context window, Mercury 2 or Kimi K3?

Kimi K3, with 1.05M tokens against 128K.

How many benchmarks is the Mercury 2 vs Kimi K3 comparison based on?

38 benchmarks have a result for both models as of October 21, 2026. Mercury 2's coding index rests on 1 independent benchmark and Kimi K3's on 2. Where only one model has a result, the page says so instead of naming a winner.

Compare next