MiniMax-M1-80k
About MiniMax-M1-80k
MiniMax-M1 is a open-weight, large-scale hybrid-attention reasoning model with 456 B parameters and 45.9 B activated per token. It natively supports 1 M-token context, lightning attention enabling 75% FLOPs savings vs DeepSeek R1 at 100 K tokens, and leverages a MoE architecture. Efficient RL training with CISPO and hybrid design yields state-of-the-art performance on long-input reasoning and real-world software engineering tasks.
Discover how MiniMax-M1-80k's 1M-token context and advanced reasoning tackle complex, real-world challenges across diverse domains.
Scientific Discovery Acceleration
Accelerate research by analyzing vast datasets, generating and verifying complex proofs, and drafting technical papers with deep, step-by-step reasoning.
Use Case Example:
"Assisted a genomics researcher by analyzing 500k lines of sequencing data to identify novel gene interactions, reducing analysis time by weeks and suggesting new experimental pathways."
Advanced Software Engineering
Beyond debugging, MiniMax-M1-80k analyzes entire codebases, identifies architectural flaws, suggests security enhancements, and optimizes performance with deep algorithmic understanding.
Use Case Example:
"Identified a critical race condition in a large-scale Python data processing pipeline by reasoning through concurrent execution paths, providing a precise fix that improved data integrity and throughput."
Deep Financial & Market Intelligence
Perform multi-step quantitative analysis on extensive financial reports and market data (1M tokens), inferring causal relationships and generating detailed, actionable strategic recommendations.
Use Case Example:
"Analyzed a target company's last five years of financial statements, market news, and regulatory filings (over 500k tokens) to produce a comprehensive M&A due diligence report, highlighting hidden risks and synergy opportunities."
Comprehensive System & Contract Auditing
Deploy AI to audit complex systems, from legal contracts to engineering schematics, by reasoning through logical dependencies, identifying inconsistencies, and flagging potential vulnerabilities or compliance issues.
Use Case Example:
"Reviewed a complex cloud infrastructure configuration (Terraform files, network policies, IAM roles) for a multi-tenant SaaS platform, identifying several security misconfigurations and compliance gaps against industry standards."
Metadata
Specification
State
Deprecated
Architecture
hybrid-attention Mixture-of-Experts (MoE)
Calibrated
Yes
Mixture of Experts
Yes
Total Parameters
456B
Activated Parameters
45.9B
Reasoning
No
Precision
FP8
Context length
131K
Max Tokens
131K
Compare with Other Models
See how this model stacks up against others.
MiniMaxAI
chat
MiniMax-M2.5
Release on: Feb 15, 2026
Total Context:
197K
Max output:
131K
Input:
$
0.2
/ M Tokens
Output:
$
1.0
/ M Tokens
MiniMaxAI
chat
MiniMax-M2.1
Release on: Dec 23, 2025
Total Context:
197K
Max output:
131K
Input:
$
0.29
/ M Tokens
Output:
$
1.2
/ M Tokens
MiniMaxAI
chat
MiniMax-M2
Release on: Oct 28, 2025
Total Context:
197K
Max output:
131K
Input:
$
0.3
/ M Tokens
Output:
$
1.2
/ M Tokens
MiniMaxAI
chat
MiniMax-M1-80k
Release on: Jun 17, 2025
Total Context:
131K
Max output:
131K
Input:
$
0.55
/ M Tokens
Output:
$
2.2
/ M Tokens
