MiniMax M2.7
Model Self-Improvement, Driving Productivity Innovation Through Technological Breakthroughs
Build a Complex Agent Harness Independently to Accomplish Highly Complex Productivity Tasks.
Model Highlights
M2.7 demonstrates excellent performance in real-world software engineering, including end-to-end project delivery, log analysis for bug hunting, code security, and machine learning tasks.
In professional office domains, we enhanced the model's domain expertise and task delivery capabilities across fields. On GDPval-AA, M2.7 achieves an ELO score of 1495, the highest among open-source models. M2.7 shows significant improvement in complex editing capabilities for Office Suite (Excel/PPT/Word), better handling multi-turn modifications and high-fidelity edits.
M2.7 possesses the ability to interact with complex environments. On 40 complex skills (>2000 Token) cases, M2.7 maintains a 97% skill adherence rate. In OpenClaw usage, M2.7 shows significant improvement over M2.5, approaching the latest Sonnet 4.6 on MMClaw evaluation.
M2.7 demonstrates excellent identity preservation and emotional intelligence. Beyond productivity use cases, it also opens space for innovation in interactive entertainment scenarios.
Benchmark
On the SWE-Pro benchmark, M2.7 scores 56.22%, nearly matching Opus's best level. This capability extends to end-to-end full project delivery (VIBE-Pro 55.6%) and deep understanding of complex engineering systems (Terminal Bench 2 at 57.0%).

M2.7 Showcases
See What M2.7 Can Do
All examples below were generated by MiniMax M2.7 in a single shot. Click cards to enlarge.
Invite friends, earn benefits
Subscribe to a Token Plan got a 10% discount, while the inviter got a 10% rebate!
Invite friends, earn benefits
Subscribe to a Token Plan got a 10% discount, while the inviter got a 10% rebate!
01 / Access Method
Quick API Integration
Two API versions: M2.7 and M2.7-highspeed with identical results but faster speed. Full automatic Cache support, no configuration needed.
02 / Access Method
For AI Coding Tools
Subscribe to the Token Plan
The price remains unchanged, while performance has significantly improved. Token Plan users now automatically benefit from higher inference speeds.
Read MoreMiniMax Agent Integration
The general Agent platform based on M2.7 is now fully open. Experience the best programming assistance and logical reasoning capabilities without any development required.
Read More













