Key Info

Together AI announced that MiniMax M3.1 Flash is now available for post-training on its platform, enabling users to run supervised fine-tuning (SFT) or reinforcement learning (RL) with their own data, rewards and environments.

Highlights

  • MiniMax M3.1 Flash can now be post-trained on Together AI
  • Supports both SFT and RL workflows
  • Users supply their own data, reward functions and environments
  • The training loop stays in the user's own Python code
  • Aimed at shaping the model for specific agentic workloads