Efficiency competition in the field of large AI models is intensifying again. MiniMax officially announced on September 27 that its latest text model, M3.1-Flash-Preview, is now fully available on the MiniMax Code platform. To thank developers for their support, the official released a major benefit on the launch day - the quota reset card, which resets all users' Token Plan quotas, and also launched multiple benefits such as free token collection.

Regarding core capabilities, M3.1-Flash-Preview is positioned as a stable productivity tool for daily development work. Whether it's regular bug fixes or complete feature development from scratch, the model can deliver reliably. It not only has the ability to deeply understand complex requirements, but also carefully handles various edge cases (boundary conditions), automatically completes regression testing, and strictly verifies the potential impact of code changes on existing functions. By organically connecting all aspects, including problem identification, code implementation, testing verification, and result delivery, the model builds an efficient and reliable development loop for developers.

image.png

In addition to the iterative upgrades of the core model features, MiniMax also announced that the limited-time "Double Points for Check-in" activity will officially start on September 28 and end on October 7. During this period, all new and existing users who log in to the MiniMax Code platform and complete daily check-ins can easily claim double free points, further reducing the cost of using the platform for developers.