Back to Models

MiniMax: MiniMax M3

Minimaxtext
9.0 / 10 Overall Rating

MiniMax M3 is a multimodal foundation model supporting text, image, and video inputs with text output. It features a 1M-token context window alongside an exceptionally high 512K output token limit. Priced at $0.30/$1.20 per million tokens, it provides high-throughput processing for long-horizon agentic workflows and code analysis at a fraction of baseline API costs.

Context Window
1049K
Knowledge Cutoff
2025
Max Output
16K
License
Proprietary

Where it Excels (Pros)

  • 1M context, massive 512K output
  • 8x cheaper than GPT-4o
  • Native text, image, video input

Limitations (Cons)

  • Reasoning lags Claude 3.5 Sonnet
  • Variable latency on ultra-large outputs

Benchmark Breakdown

Reasoning (MMLU)82%
Coding (HumanEval)78%
Mathematics (MATH)76%
Long-Context Retrieval95%

Pricing Matrix

Input cost / 1M tokens$0.30
Output cost / 1M tokens$1.20

Compare MiniMax: MiniMax M3 with other frontier models:

Related Skills, Tools & Automations