暂无简介

MiniMax-M2, a model built for Max coding & agentic workflows.

[NeurIPS 2025] The official repo of SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond

The official repo of One RL to See Them All: Visual Triple Unified Reinforcement Learning

全球首个开放权重、大规模混合注意力推理模型 MiniMax-M1，采用独特混合专家（MoE）架构结合闪电注意力机制，实现高效率推理。在复杂的软件工程、工具使用和长文本场景中表现出色，是新一代语言模型代理应对现实世界挑战的强大基础。【此简介由AI生成】

A minimal yet professional single agent demo project that showcases the core execution pipeline and production-grade features of agents.

The official GitHub Page for MiniMax

MiniMax-M1-80k模型，全球首个开放权重的大规模混合注意力推理模型，采用独特的混合专家（MoE）架构结合闪电注意力机制，能高效处理长文本输入，特别适合需要深入思考的复杂任务

MiniMax Search is an MCP (Model Context Protocol) server that provides web search and browsing capabilities.

MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model.

Explore these applications integrating MiniMax's multimodal API to see how text, vision, and speech processing capabilities are incorporated into various software.

The official repo of MiniMax-Text-01 and MiniMax-VL-01, large-language-model & vision-language-model based on Linear Attention

暂无简介

Official MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.

A collection of optimized utilities for text-to-audio processing, enhancing both training and inference workflows. This repository contains robust implementations adapted from open-source libraries.

MiniMax-M2.5开源模型，经数十万复杂环境强化训练，在代码生成、工具调用、办公自动化等经济价值任务中表现卓越。SWE-Bench Verified得分80.2%，Multi-SWE-Bench达51.3%，BrowseComp获76.3%。推理速度比M2.1快37%，与Claude Opus 4.6相当，每小时仅需0.3-1美元，成本仅为同类模型1/10-1/20，为智能应用开发提供高效经济选择。【此简介由AI生成】

A high-throughput and memory-efficient inference and serving engine for LLMs

暂无简介

暂无简介

Official MiniMax Model Context Protocol (MCP) JavaScript implementation that provides seamless integration with MiniMax's powerful AI capabilities including image generation, video generation, text-to-speech, and voice cloning APIs.