MiniMax M3.1 Flash Opens 428B Weights

This is a meaningful open-weight release: frontier-scale capacity with a much smaller active footprint, though local deployment will still demand serious infrastructure.

• –1M-token context supports long-running coding and agent workflows
• –23B active parameters improve inference efficiency versus dense 428B models
• –Space Bunny’s stealth run provided unusually large-scale real-world validation
• –Developers can inspect, fine-tune, and self-host the model instead of relying solely on MiniMax’s API
• –Hardware requirements and licensing terms remain critical details for production adoption