MiniMax M3.1 Flash

Language modelCard summarized

Frontier multimodal coding model with 1M context window and tunable thinking depth

Active (latest)Source: MiniMax M3.1 Flash launch post

Use it via API

$83.70 /month

$1.67 per 1K requests

RAG over docs: 12K in · 600 out · 50K requests/mo · 30% cached · change

Code example Compare

Where to call it

MiniMax APIMiniMax-M3.1-Flash

Supports

Image input
Streaming
Tool calling
Web search
Prompt caching

Full feature support

Good for

Enterprise workflow automation across SaaS systems (sales, marketing, ops, support, finance, HR)
Coding and software engineering at low cost
Computer use and cross-application tasks

Overview

MiniMax · 01Flash

M3.1 Flash

Frontier multimodal coding model with 1M context window and tunable thinking depth

Cost: $0.15 in
Context: 1M

Input / output: $0.15 / $0.60
Max output: 512K
Knows up to: —
Released: 19 Oct 2026

MiniMax's next-generation Flash model for agentic tasks, coding, computer use and cybersecurity. MiniMax says it is nearly three times faster in online inference than its predecessor and cuts agentic usage cost by 70%. An early version ran anonymously as "Space Bunny" and processed more than 200 trillion tokens. launch post

MiniMax-M3.1-Flash

Input / output per 1M: $0.15 / $0.60
Context window: 1M
Max output: 512K
Knows up to: —
Reasoning: Always on (low / medium / high / xhigh / max)
Weights: Closed

About this card. A model card is the maker's own account of a model: what it's for, how it was trained and tested, and where it falls short. This page condenses MiniMax's MiniMax M3.1 Flash launch post. Every claim carries its source page.