The Model Index
Up to four models

Kimi K3 vs Grok 4.5

On the index (GPT-4 = 100), Kimi K3 scores 168 and Grok 4.5 161. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to Kimi K3
Presets
Kimi K3
168152–183
WeightsOpen
ReleasedJul 2026
Price in / out$0.19 / $16
Context1.05M
Core tests taken6 of 9
Grok 4.5
161145–177
WeightsClosed
ReleasedJul 2026
Price in / out$2.0 / $6.0
Context500K
Core tests taken6 of 9

Kimi K3 leads Grok 4.5 by 7 points, ahead on 4 of 6 shared core tests, at 1.4× the price.

Test by testthird-party results

Kimi K3Grok 4.5
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)LiveBench Data AnalysisLiveBench Data AnalysisLiveBench IFLiveBench IFLiveBench LanguageLiveBench LanguageLiveBench ReasoningLiveBench ReasoningGPQA Diamond (Vals AI)GPQA Diamond (Vals AI)MMLU ProMMLU ProMMMU ProMMMU ProSimpleBenchSimpleBenchSimpleQA VerifiedSimpleQA VerifiedMATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)LiveBench MathematicsLiveBench MathematicsProofBench v1.1ProofBench v1.1Chess puzzlesChess puzzlesFrontierMath (tier 4)FrontierMath (tier 4)Surface Evolver BenchSurface Evolver BenchCODINGTerminal-Bench 4.0Terminal-Bench 4.0LiveBench Agentic CodingLiveBench Agentic CodingLiveBench CodingLiveBench CodingCode MigrationCode MigrationCyberBench v1.1CyberBench v1.1IOIIOILiveCodeBenchLiveCodeBenchProgramBenchProgramBenchSWE-benchSWE-benchVibe Code Bench v1.1Vibe Code Bench v1.1DeepSWEDeepSWEMirrorCodeMirrorCodeWeirdMLWeirdMLAGENTSAPEX-AgentsAPEX-AgentsDTBenchDTBenchFurniture assemblyFurniture assemblyLMCALMCAPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)CorpFin v2CorpFin v2EMBEMBLegalBenchLegalBenchLegal Research BenchLegal Research BenchMedCodeMedCodeMedScribeMedScribeMortgageTaxMortgageTaxTax Agent BenchTax Agent BenchTaxEval v2TaxEval v2NOVEL PROBLEMSARC-AGI-2ARC-AGI-2ARC-AGI-1ARC-AGI-1
Every result as a table
TestKimi K3Grok 4.5
Reasoning & knowledge
GPQA Diamond (Epoch AI)93.1%93.4%
LiveBench Data Analysis78.7%73.0%
LiveBench IF71.4%71.5%
LiveBench Language85.5%82.8%
LiveBench Reasoning90.7%87.2%
GPQA Diamond (Vals AI)92.9%92.9%
MMLU Pro88.0%89.2%
MMMU Pro88.1%61.8%
SimpleBench60.7%70.0%
SimpleQA Verified50.6%48.3%
Maths
FrontierMath (tiers 1-3)72.2%57.2%
AIME-style maths (OTIS mock)97.2%97.8%
LiveBench Mathematics84.4%90.8%
ProofBench v1.187.0%31.0%
Chess puzzles37.0%35.0%
FrontierMath (tier 4)39.0%24.4%
Surface Evolver Bench95.0%74.4%
Coding
Terminal-Bench 4.017.2%8.6%
LiveBench Agentic Coding62.2%56.5%
LiveBench Coding81.5%68.6%
Code Migration16.1%36.6%
CyberBench v1.175.2%71.0%
IOI48.9%40.6%
LiveCodeBench87.2%87.4%
ProgramBench2.0%0.0%
SWE-bench93.4%86.6%
Vibe Code Bench v1.185.0%69.0%
DeepSWE68.5%53.8%
MirrorCode54.4%32.2%
WeirdML82.6%46.4%
Agents
APEX-Agents47.8%56.2%
DTBench91.2%96.5%
Furniture assembly34.2%22.5%
LMCA52.7%45.2%
Professional work
Finance Agent (v2)53.1%48.4%
CorpFin v271.6%67.4%
EMB66.7%52.9%
LegalBench86.2%86.0%
Legal Research Bench46.2%38.0%
MedCode49.4%43.3%
MedScribe88.0%86.9%
MortgageTax66.3%61.8%
Tax Agent Bench33.4%24.9%
TaxEval v275.7%71.7%
Novel problems
ARC-AGI-260.4%52.6%
ARC-AGI-194.5%87.2%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG