Compare / MiniMax M2.5 vs Kimi K3

MiniMax M2.5 vs Kimi K3

Which is cheaper, which is more capable, and which does better at each kind of work, from 12 results on 3 sources that measured both models.

MiniMax M2.5
MiniMax · profile
Kimi K3
Moonshot AI · profile
Results compared
12
Shared sources
3

MiniMax M2.5 or Kimi K3? The short answer

  • MiniMax M2.5 is cheaper: $0.52 against $6.00 per million tokens, 11 times less.
  • Kimi K3 scores higher on the Artificial Analysis Intelligence Index: 43.6 against 22.8.
  • On our business benchmarks, Kimi K3 ranks higher for agents and tool use, reasoning and knowledge, writing and sticking to the facts.

At a glance

MiniMax M2.5Kimi K3
Price per million tokens$0.30 in · $1.20 out$3.00 in · $15.00 out
Blended (3 in : 1 out)$0.52$6.00
Cheapest host, blended$0.44 (Venice)$3.71 (Relace, FP4)
Intelligence Index22.843.6
Spring Prompt overall–49
Output speed, tokens a second–45
Context window204,800 tokens1,048,576 tokens
WeightsOpenOpen
Developer based inChinaChina

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.

Use caseMiniMax M2.5Kimi K3
Product listingsTurning a sparse product feed and photos into listings that can go live not measured 12 of 19
Decks from an analysisTurning a finished analysis into a deck you could present as it is not measured 8 of 18
Marketing planningPlanning a year of ad spend without overspending not measured 8 of 18
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 132 of 161 32 of 161
Professional workReal tasks from banking, consulting and law, business documents and freelance projects not measured 24 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 80 of 195 15 of 195
WritingWhat people prefer in blind comparisons, and judged writing quality 112 of 187 11 of 187
Sticking to the factsSummarising without inventing things, and factual answers 103 of 153 31 of 153
Speed under pressureGood decisions against a real clock (fast chess) not measured 18 of 21

Every shared result

BenchmarkMetricMiniMax M2.5Kimi K3
Arena (formerly LMArena)reportedOverallArena rating 1,3861,472
Arena (formerly LMArena)reportedBusiness, management and financeArena rating 1,4001,489
Arena (formerly LMArena)reportedCreative writingArena rating 1,3571,458
Arena (formerly LMArena)reportedExpert promptsArena rating 1,4251,533
Arena (formerly LMArena)reportedInstruction followingArena rating 1,3811,486
Arena (formerly LMArena)reportedOverallArena rating 1,3911,488
Arena (formerly LMArena)reportedWriting, literature and languageArena rating 1,3691,475
Artificial AnalysisreportedArtificial Analysis Intelligence Indexindex score 22.843.6
Artificial AnalysisreportedGPQA Diamond% of questions 84.8%93.5%
Artificial AnalysisreportedHumanity's Last Exam% of questions 20.5%46.9%
Artificial AnalysisreportedLong-context reasoning (AA-LCR)% of questions 73.3%88.7%
Vending-Bench 2reportedMoney after a yearUS dollars $-23.16$5,165.03

Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.

Quick answers

Is MiniMax M2.5 better than Kimi K3?

It depends on the task. Among models available today, on our benchmarks and licensed sources, Kimi K3 ranks higher for agents and tool use, reasoning and knowledge, writing and sticking to the facts.

Which is cheaper, MiniMax M2.5 or Kimi K3?

MiniMax M2.5 costs $0.52 per million tokens (three input to one output), against $6.00 for Kimi K3.

Run MiniMax M2.5 and Kimi K3 on your own prompt

Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.