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
Run agent, coding, computer-use and security workloads on MiniMax M3.1 Flash — 200+ TPS output at a fraction of previous token spend.
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
Where MiniMax M3.1 Flash Earns Its Place in Production
Successor to the anonymously run Space Bunny system, MiniMax M3.1 Flash has been measured on enterprise automation, software engineering, desktop operation and security analysis.
- Multi-Step Agent AutomationSits near the top of AutomationBench-AA and posts 76.2 on BankerToolBench, carrying full SaaS processes through sales, marketing, ops, support, finance and HR.
- Software Engineering DepthScores 73.7% on DeepSWE v1.1 — reading unfamiliar code, shipping features, patching defects and running its own tests over long task chains.
- Desktop Control and Security ReviewHits 61.7 on OSWorld V2 and 29.6 on Agents' Last Exam for live-environment operation, with 82% on CyberGym for finding and confirming vulnerabilities.
Getting MiniMax M3.1 Flash into Your Stack
Three moves take you from a first call to dependable, production-level agent tasks.
Under the Hood of MiniMax M3.1 Flash
Quicker inference, predictable per-token costs, and measured performance in automation, software work, desktop control and security engineering.
Sparser Serving, Quicker Replies
Heavier sparsity in the serving layer pushes routine output past 200 TPS with a P50 above 170 TPS — roughly a threefold gain in online inference speed.
Predictable, Budget-Friendly Rates
At $0.15 per 1M input, $0.60 per 1M output and $0.015 per 1M cached input, total agent running costs land about 70% under the prior generation.
End-to-End SaaS Process Runs
A leading AutomationBench-AA position plus 76.2 on BankerToolBench show multi-step planning, finance compliance handling and linked tool calls inside live SaaS systems.
Real Software Delivery on DeepSWE
A 73.7% DeepSWE v1.1 result places it among top models on demanding engineering jobs needing independent code reading, feature work, bug repair and test checks.
Hands-On Desktop Operation
With 61.7 on OSWorld V2 and 29.6 on Agents' Last Exam — leading Chinese models on both — it drives real environments, operates tools and returns finished output.
Security Work Without Heavy Spend
An 82% CyberGym score, checked by in-house experts, spans code auditing, patch validation, pen testing, malware triage and reverse engineering at modest cost.
MiniMax M3.1 Flash: Your Questions Answered
Straight answers on MiniMax M3.1 Flash pricing, output speed, benchmark scores and how to get access.
So what is MiniMax M3.1 Flash, exactly?
A next-generation Flash model from MiniMax, grown out of the anonymous Space Bunny system that handled more than 200 trillion tokens; it moves ahead in agent work, coding, desktop control and cybersecurity.
What does a million tokens cost?
Rates are flat — $0.15 per 1M input, $0.60 per 1M output and $0.015 per 1M cached input — leaving agent workloads roughly 70% less expensive than the earlier model.
Where does the near-3x speed gain come from?
Extra sparsity throughout the serving stack: routine throughput reaches 200+ TPS while the median (P50) holds above 170 TPS.
Which benchmark scores support these numbers?
A near-top AutomationBench-AA rank, 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 start using it?
Use the MiniMax API on platform.minimax.io, an M Plan subscription, or MiniMax Code — the desktop build comes from agent.minimax.io/download; open weights arrive later.
How is Flash different from Pro?
Flash is MiniMax's continuing study in balancing efficiency against capability; Pro, built to stretch the frontier of raw intelligence, lands later.
Put MiniMax M3.1 Flash to Work
Spin up the MiniMax API, subscribe to an M Plan or grab MiniMax Code — strong agent, coding, desktop and security performance at about 70% less spend and close to 3x the speed.