MiniMax

MiniMax M3

Released 2026-05-31

MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,...

You send / 1M

$0.300

tokens in

It replies / 1M

$1.20

tokens out

Memory

1.05M

tokens it can read at once

Typical request

$0.0010

1,000 words in, 400 out

Data privacy

No training or retention: Yes. At least one provider serves it with zero data retention.

ProviderYou send / 1MIt replies / 1MUptimePrivacy policy
DeepInfra · fp8$0.224$0.88090%Read
CoreWeave · fp4$0.230$0.960100%Read
DeepInfra · fp8$0.280$1.1099%Read
Venice · fp8$0.300$1.2098%Read
Together$0.300$1.20100%Read
Parasail · fp8$0.300$1.2097%Read
Novita · fp8$0.300$1.20100%Read
ModelRun · fp8$0.300$1.3098%Read
Mara$0.600$2.40100%Read

Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.

What it can do

  • Reads: text, image, video
  • Writes: text
  • Reasoning: Yes
  • Tool calling: Yes
  • Longest reply: 512K tokens
  • Repeated text is billed at $0.060 per 1M (prompt caching)
  • Intelligence score: 29.2 / 100 (Artificial Analysis)

Similar models

Appears in

Estimate your monthly cost with MiniMax M3 →