Back to MiniMax M3

MiniMax M3

MiniMax M3 is a multimodal foundation model with 1M token context, native multimodal understanding, and MiniMax Sparse Attention (MSA) for efficient long-context inference.

minimax/minimax-m3
STABLEGet Started
Streaming
Vision
Tools
Reasoning
JSON Output

Select Provider

MiniMax Pricing for MiniMax M3

View detailed pricing and capabilities for this provider.

MiniMax
Context: 1.0M
Input
$0.6
/M tokens
Cached
$0.12
/M tokens
Output
$2.4
/M tokens
Get Started