Skip to main content
MI

MiniMax-M3

minimax/minimax-m3

MiniMax multimodal model for long-context coding, perception, and agent planning

Compare
Context window
512K
Input
Image · Text · Video
Output
Text
Released
2026-06-01
Max output
128K
Model parameters
427B
Weights
Open
License
minimax-community
Capabilities
Attachments, Open weight, Reasoning, Temperature, Tool call

Pricing

IN

$0.30

USD / 1,000,000 token

prompt: <=512000 input tokens

OUT

$1.20

USD / 1,000,000 token

prompt: <=512000 input tokens

Cache read

$0.06

USD / 1,000,000 token

prompt: <=512000 input tokens

Performance

Overall capability

44.7/100

Coding
39.3
Reasoning
42.6
Professional work
41.6
Knowledge & language
51.0
Terminal-Bench Science · AAMiniMax-M3ReasoningArtificial Analysis2026-11-05
SciCodeMiniMax-M3CodingArtificial Analysis2026-11-05
Vibe Code Bench 1–100CodingVals AI2026-11-05
AA-OmniscienceMiniMax-M3Knowledge & languageArtificial Analysis2026-11-05
Chess PuzzlesReasoningEpoch AI2026-11-05
Terminal-Bench 4.0 (Vals rerun)CodingVals AI2026-11-05
Finance AgentProfessional workVals AI2026-11-05
CritPtMiniMax-M3ReasoningArtificial Analysis2026-11-05

Providers

Charm Hyperminimax-m3Openai chat--
Cortecsminimax-m3Openai chat--
CrossModelminimax/minimax-m3Openai chat-
EmpirioLabs AIminimax-m3Openai chat--
Kenariminimax-m3Openai chat---
Kilo Gatewayminimax/minimax-m3Openai chat--
LLM Gatewayminimax-m3Openai chat--
Merge Gatewayminimax/minimax-m3Openai chat--
MiniMax (minimax.io)MiniMax-M3Custom--
MiniMax (minimaxi.com)MiniMax-M3Custom--
NanoGPTminimax/minimax-m3Openai chat--
Ollama Cloudminimax-m3Openai chat-----
OpenCode Gominimax-m3Openai chat--
OpenCode Zenminimax-m3Openai chat--
OpenRouterminimax/minimax-m3Openai chat--
TensorXminimax/minimax-m3Openai chat--
Vercel AI Gatewayminimax/minimax-m3Openai chat--
WaferMiniMax-M3Openai chat-
ZenMuxminimax/minimax-m3Openai chat---

ModelRegistry · Catalog updated 2026-11-05