The Model Index
Up to four models

Gemini 4 Argon vs Kimi K3.1 Pro

On the index (GPT-4 = 100), Gemini 4 Argon scores 191 and Kimi K3.1 Pro 179. 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 Gemini 4 Argon
Presets
Gemini 4 Argon
191176–207
WeightsClosed
ReleasedSep 2026
Price in / out$2.0 / $10
Context1.05M
Core tests taken6 of 9
Kimi K3.1 Pro
179163–195
WeightsOpen
ReleasedOct 2026
Price in / out$0.24 / $8.4
Context1.05M
Core tests taken4 of 9

Gemini 4 Argon leads Kimi K3.1 Pro by 12 points, ahead on 3 of 4 shared core tests, at 1.8× the price.

Test by testthird-party results

Gemini 4 ArgonKimi K3.1 Pro
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)SimpleQA VerifiedSimpleQA VerifiedMATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)ProofBench v1.1ProofBench v1.1Chess puzzlesChess puzzlesFrontierMath (tier 4)FrontierMath (tier 4)CODINGTerminal-Bench 4.0Terminal-Bench 4.0IOIIOIVibe Code Bench v1.1Vibe Code Bench v1.1AGENTSAPEX-AgentsAPEX-AgentsBioMysteryBenchBioMysteryBenchSRE BenchSRE BenchTerminal-Bench-ScienceTerminal-Bench-ScienceDTBenchDTBenchFurniture assemblyFurniture assemblyLMCALMCAMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKHLAB-Vals-1.1HLAB-Vals-1.1Legal Research BenchLegal Research BenchTax Agent BenchTax Agent BenchNOVEL PROBLEMSMysteryMechanismMysteryMechanism
Every result as a table
TestGemini 4 ArgonKimi K3.1 Pro
Reasoning & knowledge
GPQA Diamond (Epoch AI)96.4%94.2%
SimpleQA Verified76.0%62.2%
Maths
FrontierMath (tiers 1-3)86.0%78.3%
AIME-style maths (OTIS mock)100.0%100.0%
ProofBench v1.199.0%91.0%
Chess puzzles59.0%45.0%
FrontierMath (tier 4)78.0%65.8%
Coding
Terminal-Bench 4.057.6%63.6%
IOI100.0%81.9%
Vibe Code Bench v1.191.9%84.9%
Agents
APEX-Agents82.4%65.5%
BioMysteryBench76.3%75.6%
SRE Bench44.3%27.1%
Terminal-Bench-Science44.3%31.4%
DTBench97.6%93.3%
Furniture assembly72.5%55.8%
LMCA63.7%57.6%
Mystery game puzzles64.0%42.0%
Professional work
HLAB-Vals-1.127.1%11.7%
Legal Research Bench54.8%37.5%
Tax Agent Bench49.2%24.9%
Novel problems
MysteryMechanism45.5%26.1%

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

This page as a chartDownload PNG