MiniMax M2.7
FrontierMiniMax•Released on 2026-03-18
MiniMax's self-evolving AI model with breakthrough agent capabilities. Demonstrates 30-50% autonomous RL research workflow. Excels at software engineering (SWE-Pro 56.22%), professional office tasks (GDPval-AA Elo 1495), and complex tool-calling with 97% skill adherence. Features significantly reduced hallucination (34% rate) and 20% fewer tokens than competitors.
83
Overall Score
Core Specs
200K
Context Window
32K
Max Output
✓
Reasoning
✗
Open Source
Multimodal Support
text
Scenario Scores
User Feedback Highlights
Based on community feedback. Hover to see original reviews.
+ GDPval-AA Elo 1495 (highest among accessible models)+ Self-evolving RL capabilities (30-50% autonomous workflow)+ MLE Bench Lite 66.6% medal rate (ties Gemini 3.1)+ Extremely cheap ($0.30/1M input, $1.20/1M output)+ 20% fewer output tokens than competitors− Lower BridgeBench score than M2.5− Less known in Western markets− Smaller ecosystem than Claude/GPT+ Low hallucination rate (34% vs 46% Claude Sonnet 4.6)− Documentation mainly in Chinese+ 97% skill adherence on complex tasks− Proprietary model (weights not open source)
Sentiment:👍 72%😐 22%👎 6%
Pros & Cons
Pros
- +Self-evolving RL capabilities (30-50% autonomous workflow)
- +Extremely cheap ($0.30/1M input, $1.20/1M output)
- +Low hallucination rate (34% vs 46% Claude Sonnet 4.6)
- +97% skill adherence on complex tasks
- +20% fewer output tokens than competitors
- +MLE Bench Lite 66.6% medal rate (ties Gemini 3.1)
- +GDPval-AA Elo 1495 (highest among accessible models)
Cons
- −Proprietary model (weights not open source)
- −Less known in Western markets
- −Documentation mainly in Chinese
- −Lower BridgeBench score than M2.5
- −Smaller ecosystem than Claude/GPT
Reliability
Incidents (30d)0
Pricing
Input (per 1M tokens)$0.30
Output (per 1M tokens)$1.20
Free trial available
Updated on 2026-03-19
Get Started
1Visit the provider's website
2Create an account
3Start using the model
Compare with Others
Benchmarks
userRating8.6%
aiIndex50%