Atlas

Compare Models

Kimi K3.1 Pro vs MiniMax M3.1 Flash

Kimi K3.1 Pro leads by 4.0 capability points (≈95% interval +1.5 to +6.5): a clear lead over MiniMax M3.1 Flash · P(higher) ≈ 100%

Wins 61 of 70 shared benchmarks (7 lost, 2 tied)

Capability vs. relative cost
Kimi K3.1 ProMiniMax M3.1 FlashClosed weightsOpen weights

Each line runs through one model’s priced reasoning settings, cheapest to most expensive; grey points are other models’ priced settings in the same range. Results whose reasoning effort the source didn’t report are left out wherever a model has named effort levels. Capability and relative cost are the published estimates (relative cost is cost per task as a multiple of GPT-5 (high)); hover a setting, or use the arrow keys, to set it beside the other model’s same setting.

Kimi K3.1 ProvsMiniMax M3.1 Flashcapability vs. relative costData Nov 15, 2026Village LLM Atlasatlas.village.dev

Benchmark wins61 – 7 · 70 shared, 2 tied

Comparing matching setups only (15 benchmarks excluded) · suites counted once
← Kimi K3.1 ProMiniMax M3.1 Flash →

Kimi K3.1 ProvsMiniMax M3.1 Flashwins by categoryData Nov 15, 2026Data: Artificial Analysis, Vals AI, Mercor and 13 othersVillage LLM Atlasatlas.village.dev

Margins

Each benchmark scaled 0–100% (its declared range, else the field’s range); Δ is MiniMax M3.1 Flash minus Kimi K3.1 Pro.

AA-PersonalityBench Transparency indexChat & Writing88.631.5−57.7
AA-PersonalityBench Warmth indexChat & Writing58.112.4−46.2
AA-PersonalityBench Imagination indexChat & Writing54.312.4−42.3
Terminal-Bench 4.0 (Artificial Analysis)Code67.225.8−41.4
Terminal-Bench 4.0Code63.625.3−38.4
Legal Research Bench HealthLegal73.336.7−36.7
AA-PersonalityBench Expressiveness indexChat & Writing65.729.6−36.5
NoCoT-Bench NCKI (kspine v3)General QA11665.9−34.9
Terminal-Bench-Science 0.1Science34.33.3−31.0
VoxelBench Text RatingGames22161810−28.9
BuseyBench SVGMultimodal6.94.2−27.2
τ³-Banking (Mercor run, bm25_grep retrieval)Agents59.836.1−23.7
ALE-BenchCode21321495−22.6
DAXBench v2Code86.263.8−22.3
Medical Long Context Reasoning (MLCR-AA, October 2026 methodology)Health & Medicine38.316.7−21.7
Visa Agentic Commerce BenchmarkAgents53.333.3−20.0
CritPt v2 (agentic, avg@5)Science70.651.4−19.1
NoCoT-Bench NCRI 15.2General QA10368.4−18.9
GDPval-AA v2.1Professional Work18611465−17.4
Artificial Analysis Intelligence Index v5.0.1Indexes46.529.1−17.4
Artificial Analysis Intelligence Index v5.0Indexes46.028.9−17.1
GBENCH One-shot Coding GScoreGames58.241.1−17.1
VIALS-AA (pass^3)Multimodal45.329.8−15.5
Legal Research Bench Administrative/RegulatoryLegal56.140.9−15.2
Agents' Last Exam v1.1 (ALE v1.1)Agents43.429.5−13.9
Artificial Analysis Intelligence Index v4.3.2Indexes55.142.7−12.4
Legal Research Bench Weighted PassLegal71.483.1+11.6
Code Migration v1.1 (Vals Index v2.2 subset)Code59.748.2−11.5
Legal Research Bench Civil LitigationLegal29.640.7+11.1
Vibe Code Bench v1.1Code84.973.9−11.0
AA-Briefcase v1.1 EloAgents16741474−10.9
GDP.pdf (Artificial Analysis, All-pass)Professional Work31.420.8−10.6
MKA (Artificial Analysis)Professional Work43.933.5−10.4
VIALS-AA (average of 3)Multimodal60.049.7−10.4
AA-PersonalityBench Informativeness indexChat & Writing31.541.0+9.6
Legal Research Bench Business/CommercialLegal27.937.2+9.3
Design Arena Overall Code CategoriesCode13531297−9.3
Humanity's Last Exam DiamondGeneral QA37.328.1−9.2
FrontierCode 1.1 Main (Artificial Analysis)Code47.138.6−8.5
Artificial Analysis Progress Index v1Indexes341318−8.1
APEX-Agents 1.1Professional Work75.968.1−7.8
AA-PersonalityBench Organization indexChat & Writing82.975.3−7.7
SciCode 1.0.1 (Artificial Analysis)Code60.853.4−7.4
AA-Omniscience IndexIndexes29.515.6−7.0
AA-LCR v1.1Search86.379.7−6.7
FrontierML ($200 wallet)Agents29.623.2−6.4
Artificial Analysis Openness IndexIndexes33.327.8−5.6
MMMU ProMultimodal83.979.1−4.7
FrontierCode 2.0 (Artificial Analysis)Code25.520.8−4.6
Humanity's Last Exam (Text-Only)General QA52.548.0−4.5
HLAB-Vals-1.1Legal11.77.5−4.2
AA-PersonalityBench Conviction indexChat & Writing65.761.9−3.8
AA-PersonalityBench Slop indexChat & Writing21.925.8+3.8
Access to Justice BenchmarkLegal69.573.3+3.8
LiveBench 2.0Agents76.773.1−3.6
Frontier3DBenchCode56.052.5−3.4
VIALS-AA (pass@3)Multimodal73.970.8−3.1
Finance Agent v2.1Finance & Accounting61.558.8−2.7
Legal Research Bench Constitutional/Civil RightsLegal31.628.9−2.6
APEX-Accounting 1.1Finance & Accounting62.460.0−2.4
CritPt (avg@5)Science24.622.3−2.3
Excel Modeling Benchmark v1.1Finance & Accounting70.067.8−2.2
Harvey LAB (Mercor run, undisclosed ~213-task cohort)Legal13.811.9−1.9
Legal Research Bench CriminalLegal28.126.3−1.8
Tax Agent BenchFinance & Accounting63.962.7−1.2
ErdosBenchMath3.03.0+1.0
AutomationBench-AA 1.0.6Agents73.372.5−0.8
Vals Finance IndexIndexes56.055.9−0.1
Legal Research Bench FamilyLegal13.613.60.0
Legal Research Bench ImmigrationLegal18.218.20.0

Kimi K3.1 ProvsMiniMax M3.1 Flashnormalised score marginsData Nov 15, 2026Data: Artificial Analysis, Vals AI, Mercor and 13 othersVillage LLM Atlasatlas.village.dev

Win counts use each benchmark’s declared score direction. Represented suite slices and the derived Capability score are excluded from headline totals. Missing or incompatible setups are excluded, never counted as losses.
Loading Atlas data…