FrontierThe story, in brief

MiniMax Just Open Sourced MiniMax M2.7: A Self-Evolving Agent Model that Scores 56.22% on SWE-Pro and 57.0% on Terminal Bench 2

56.22% on SWE-Pro. MiniMax just open-sourced M2.7, the first model that builds itself.

Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI agents and the coordination of work.AI illustration by KeyNews
The KeyNews take

Why it matters

MiniMax M2.7 represents a capability leap in open-source reasoning models with competitive SWE benchmarks, and introduces a novel self-evolving training paradigm that could reshape how foundation models are developed and distributed.

The key facts

6 to know
  1. MiniMax M2.7 open-sourced on Hugging Face

  2. 56.22% on SWE-Pro benchmark

  3. 57.0% on Terminal Bench 2

  4. First model with active participation in own development cycle

  5. Originally announced March 18, 2026

  6. Self-evolving agent architecture

Go to the source

MarkTechPostmarktechpost.com

Publisher excerpt: MiniMax has officially open-sourced MiniMax M2.7, making the model weights publicly available on Hugging Face. Originally announced on March 18, 2026, MiniMax M2.7 is the MiniMax’s most capable open-source model to date — and its first model to actively participate in its own development cycle, a…
Read original report
Back to today's editionMore frontier news

Keep reading

Related stories

More from Frontier