None
None
Long Story Video Skill
Turn your idea into 10s–10min video—AI writes scripts, prompts, and generates footage automatically.
Ads Video Skill
Generate professional ads and sales videos—AI auto-generates scripts, prompts, and footage.
3D Science Explainer Video Skill
Convert scientific concepts into stunning 3D explain animations
Feedback
freeTrialImage.bannerPity
freeTrialImage.upgradeUnlock
- ✓freeTrialImage.benefitHd
- ✓freeTrialImage.benefitWatermark
- ✓freeTrialImage.benefitUnlimited
minimax m3.1 flash
Put MiniMax M3.1 Flash behind your agents, coding jobs and computer-use tasks: 200+ TPS, low flat token rates and strong benchmark scores.
All Tools
Discover our comprehensive AI-powered animation toolkit
MiniMax H3
MiniMax H3 AI Video Generator
Seedance 2.5
The Future of AI Video Is Here.
Seedance 2.0
The Future of AI Video Is Here.
Veo3.1
Create Stunning Videos with Veo3.1
FLUX 3 Video Generator
Kling 3.0
Next-Gen AI Video Generator
Grok Video Generator
Create Videos from Text or Images with AI
MiniMax H3 video generator
MiniMax H3 AI Video Generator
MiniMax M3.1 Flash: Faster, Leaner and Built for Agent Work
Successor to the quietly deployed Space Bunny system, MiniMax M3.1 Flash has been measured on enterprise automation, software engineering, computer operation and security engineering.
- Automation That Runs End to EndIt sits near the top of AutomationBench-AA and posts 76.2 on BankerToolBench, carrying SaaS processes through sales, marketing, operations, support, finance and HR without hand-holding.
- Software Engineering DepthA 73.7% score on DeepSWE v1.1 reflects real repository work: reading unfamiliar code, shipping features, patching bugs and proving fixes with tests over long step chains.
- Real Desktops and Security Work61.7 on OSWorld V2 and 29.6 on Agents' Last Exam show it driving live environments, while 82% on CyberGym covers finding and confirming vulnerabilities.
Three Steps to Running MiniMax M3.1 Flash
From the first API call to production-grade agent tasks in three quick steps.
What You Get with MiniMax M3.1 Flash
Quick inference, predictable token costs and measured strength in automation, software work, computer operation and security engineering.
Sparser Serving, Quicker Answers
A sparser serving stack pushes routine output past 200 TPS with a median above 170 TPS — roughly a threefold gain in online inference speed.
Predictable, Budget-Friendly Rates
At $0.15 per million input tokens, $0.60 per million output and $0.015 per million cached input, agent workloads cost about 70% less than on the prior model.
Hands-On SaaS Execution
Its AutomationBench-AA standing and 76.2 on BankerToolBench come from multi-step planning, compliance checks and chained tool use inside live SaaS systems.
Repository-Level Coding
73.7% on DeepSWE v1.1 places it among top models for demanding engineering: understanding codebases, building features, resolving bugs and validating with tests.
Operating Real Desktops
Scores of 61.7 on OSWorld V2 and 29.6 on Agents' Last Exam, the best among Chinese models on both, show it working inside real environments with tools in hand.
Security Engineering Backup
An 82% CyberGym result, confirmed by in-house expert review, spans code auditing, patch work, penetration testing, malware triage and reverse engineering — all at low cost.
MiniMax M3.1 Flash: Your Questions Answered
Straight answers on MiniMax M3.1 Flash pricing, output speed, benchmark results and how to get access.
So what is MiniMax M3.1 Flash, exactly?
It is the newest Flash release from MiniMax, grown out of the unnamed Space Bunny system that handled more than 200 trillion tokens, and it advances on agentic work, coding, computer operation and cybersecurity.
What are the per-million-token rates?
Pricing is flat: $0.15 per million input tokens, $0.60 per million output tokens and $0.015 per million cached input, which leaves agent workloads around 70% cheaper than before.
Why is inference nearly three times faster?
Sparser serving is the driver: routine throughput now exceeds 200 TPS, and the median (P50) figure holds above 170 TPS.
Which benchmarks support these numbers?
It ranks near the top of AutomationBench-AA and posts 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam and 82% on CyberGym.
How do I get access?
Use the MiniMax API on platform.minimax.io, sign up for M Plan, or grab the MiniMax Code desktop build from agent.minimax.io/download; open weights arrive later.
How is Flash different from Pro?
Flash keeps testing how far efficiency and capability can be balanced; Pro, built to push raw intelligence further, lands afterwards.
Start Shipping with MiniMax M3.1 Flash
Grab the MiniMax API, choose an M Plan subscription or install MiniMax Code, and get top-tier agent, coding, computer-use and security results at about 70% less spend and roughly 3x the speed.