modelgrep
M

MiniMax models

8models tracked · ranked by intelligence, speed & price

Quick answer · Updated June 2026

MiniMax's smartest model is MiniMax M3 (54.7 on the Intelligence Index), its fastest is MiniMax M2.7 at 261 tokens/sec, and its cheapest is MiniMax M2.5 at $0.150 per million input tokens. All 8 MiniMax models are compared below by intelligence, speed, latency, context and price.

Smartest MiniMaxFastest MiniMaxCheapest MiniMaxSmall & fast MiniMaxMiniMax for codingMiniMax reasoning
Smartest
54.7
minimax/minimax-m3
Fastest
261 t/s
minimax/minimax-m2.7
Cheapest
$0.150/M
minimax/minimax-m2.5
ModelIntelIn $/MContext
minimax-m3
ReasoningToolsJSON+1
54.7$0.3001.0M
minimax-m2.7
ReasoningToolsJSON
49.6$0.250205K
minimax-m2.5
ReasoningToolsJSON
41.9$0.150205K
minimax-m2.1
ReasoningToolsJSON
39.4$0.290205K
minimax-m2
ReasoningToolsJSON
36.1$0.255205K
minimax-m2-her
—$0.30066K
minimax-m1
ReasoningTools
—$0.4001M
minimax-01
Vision
—$0.2001.0M