@heyshrutimishra: MiniMax just did something none of the big labs have pulled off. One plan. CLI, API, Agent. Every model: M2.7, video, v…
Summary
MiniMax launches a unified plan covering CLI, API, and Agent with shared credits across all models (M2.7, video, voice, music) and open-sources the entire stack including Teams and Mavis. Head of R&D hints at upcoming M3 model.
View Cached Full Text
Cached at: 05/17/26, 03:26 AM
MiniMax just did something none of the big labs have pulled off.
One plan. CLI, API, Agent. Every model: M2.7, video, voice, music. Credits shared across everything.
And they’re open-sourcing the whole stack. Teams, Mavis, all of it.
And this is just M2.7. Their Head of R&D already hinted M3 is up next.
Can’t wait to try all of this on one plan. Manging too many tools and subscription is a huge hassle…. Hoping upcoming M3 launch will be a beast.
MiniMax_Agent (@MiniMaxAgent): Upgrading Agent — here’s what’s new 🧵
🤝 Agent Teams Multi-agent, multi-stage parallel execution on desktop. Adversarial quality gates between phases. Spin up agents across roles — run your own one-person company.
🧠 Mavis — your personal AI chief of staff Lark, WeChat,
Similar Articles
@PrajwalTomar_: Everyone's sleeping on MiniMax. Again. They just shipped M3. The first open-weights model to combine frontier coding, 1…
MiniMax released M3, an open-weights model combining frontier coding, 1M context, and native multimodality, offering comparable performance to Opus at a fraction of the cost.
The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence
The MiniMax-M2 series introduces Mixture-of-Experts language models that achieve high performance on agentic tasks with minimal activated parameters (9.8B per token out of 229.9B total), leveraging agent-driven data pipelines, a scalable RL system called Forge, and a checkpoint that takes early steps toward self-evolution.
MiniMaxAI/MiniMax-M2.7
MiniMaxAI releases MiniMax-M2.7, an open-weight model featuring self-evolution capabilities, advanced agent team support, and strong performance on software engineering benchmarks (56.22% on SWE-Pro, 66.6% medal rate on MLE Bench Lite), with notable applications in production incident recovery and professional work tasks.
MiniMax M3 (2 minute read)
MiniMax introduces M3, the first open-weights model to combine coding, agentic, and multimodal capabilities with up to 1M context via sparse attention.
Testing MiniMax M2.7 via API on three real ML and coding workflows
A developer tests the MiniMax M2.7 model via its API on three practical machine learning and coding workflows, evaluating its performance.