Use Minimax M2 in Cursor: Quick Setup Guide

Stop struggling with slow or overpriced AI coding models. MiniMax M2 delivers fast reasoning and reliable agentic performance for real coding work, and you can use it directly inside Cursor.

This step-by-step guide shows you exactly how to use MiniMax M2 in Cursor using Novita AI’s OpenAI-compatible API, plus configuration tips to get the best speed, accuracy, and workflow efficiency.

What Makes Cursor IDE Essential for Developers in 2025?

Cursor is a next-generation AI-driven coding environment built for developers. Its adaptive retrieval models deeply understand your project’s structure, reducing the need to manually provide context.

Core Capabilities of Cursor IDE

• Intelligent Code Completion: Intelligently predicts and completes code based on your project’s real-time context.
• Multi-Model Flexibility: Supports various AI providers, letting you choose the best model for each development need.
• Conversational Assistance: Offers an integrated chat interface for instant help, explanations, and code generation.
• Agentic Reasoning: Handles complex, multi-step coding tasks through reasoning-driven “agent mode.”
• Seamless Testing Integration: Works effortlessly with popular testing tools and frameworks for streamlined validation.

What is Minimax M2?

MiniMax M2 is a next-generation lightweight large language model designed for coding and agentic AI workflows. Built on a Mixture-of-Experts (MoE) architecture with 230 billion total parameters and 10 billion active per token, it delivers exceptional efficiency without compromising intelligence. MiniMax M2 combines fast inference, cost-effective deployment, and robust reasoning, enabling developers to execute complex, multi-step tasks with precision.

Core Capabilities

• Superior Intelligence: MiniMax-M2 achieves top-tier general intelligence across domains such as mathematics, science, coding, and instruction following, as well as agentic tool use. Its overall composite score places it at the forefront of all open-source large language models.
• Advanced Coding: Purpose-built for complete developer workflows, MiniMax-M2 handles multi-file edits, iterative run-fix loops, and test-driven repairs with exceptional consistency. Its strong results on Terminal-Bench and Multi-SWE-Bench–style tasks highlight real-world effectiveness within terminals, IDEs, and CI pipelines.
• Agentic Capability: MiniMax-M2 demonstrates advanced planning and execution across extended toolchains involving shells, browsers, retrieval tools, and code runners. In BrowseComp-style evaluations, it reliably identifies hard-to-find sources, preserves traceable reasoning, and recovers gracefully from execution errors.
• Efficient Architecture: Featuring 230 billion total parameters with 10 billion active per token, MiniMax-M2 offers outstanding efficiency, reducing latency, cost, and resource demand while maintaining high throughput. Its design fits perfectly with the current move toward deployable, high-performance models optimized for coding and agentic applications.

Cursor Pro Pricing and Subscription Requirements

Attention: Users without a Cursor Pro subscription cannot connect personal API keys to access advanced AI coding features. Both Agent Mode and the Edit function are exclusively available to Pro subscribers.

Ask Mode (Basic AI Assistance)

• Enables conversational AI support for exploring and understanding your codebase.
• Provides intelligent search and query capabilities across files.
• Operates in a read-only environment—no direct code modifications.
• Best suited for learning, code comprehension, and safe experimentation.

Agent Mode (Advanced AI Coding)

• Unlocks autonomous coding and multi-step task execution.
• Navigates your codebase independently to implement changes.
• Plans, edits, and validates complex modifications with minimal user input.
• Supports tool calling, enabling full-cycle workflow automation.

Why Use Minimax M2 in Cursor?

Using MiniMax M2 in Cursor gives developers a faster, more affordable, and deeply integrated AI coding experience. Cursor’s IDE provides full project awareness, while MiniMax M2 delivers strong reasoning and low-latency responses through its lightweight MoE architecture. This pairing lets you handle complex refactors, dependency fixes, and multi-file edits without switching contexts or paying closed-model costs. Compared with heavier models, MiniMax M2 maintains code quality while cutting response time—making it an ideal choice for high-frequency coding, automated debugging, and agentic workflows inside Cursor.

How to Use Minimax M2 in Cursor

Prerequisite: Get an API Key

Novita AI provides Minimax
APIs with 204K context and costs of $0.3/1M input and $1.2/1M output, empowering MiniMax M2 to perform at its highest potential in agentic coding tasks.

Novita AI

Step 1: Create your Account

Sign up for a Novita AI account. Once you’re logged in, you’ll begin by generating an API key, which you’ll use to authenticate requests.m.

Step 2: Generate your API Key

Navigate to Key Management and select Add New Key. This key serves as your request credential. It will only be displayed once, so make sure to copy it immediately and store it somewhere safe—you’ll need it for the steps that follow.

Step 3: Verify Model Access

• Select a Model Name: You’ll need to copy the model name you want to use from Novita AI’s Model Library.
• API Endpoint: https://api.novita.ai/openai
• Compatibility: Full OpenAI API standard support

Top 7 AI Coding Models Available on Novita AI Platform

Model	Context Window	Best For	Model ID
Minimax M2	204.8k	Fast agentic coding workflows	minimax/minimax-m2
GLM 4.6	204.8k	General coding excellence	zai-org/glm-4.6
Qwen3-Coder 480B	262k	Specialized coding tasks	qwen/qwen3-coder-480b-a35b-instruct
DeepSeek V3.1	131k	Complex problem solving	deepseek/deepseek-v3.1
Kimi K2	262k	Large codebase analysis	moonshotai/kimi-k2-0905
OpenAI GPT OSS 120B	131k	OpenAI alternative	openai/gpt-oss-120b
Gemma 3 12B	131k	Visual + code tasks	google/gemma-3-12b-it

Complete Cursor Installation and Setting Guide

Step 1: Install and Activate Cursor

• Download the newest version of Cursor IDE from cursor.com
• Subscribe to the Pro plan to enable API-based features
• Open the app and finish the initial configuration

Step 2: Access Advanced Model Settings

• Open Cursor Settings (use Ctrl + F to find it quickly)
• Navigate to the “Models” tab in the left menu
• Find the “API Configuration” section

Step 3: Configure Novita AI Integration

• Go to the API Keys settings.
• Enable both OpenAI API Key and Override OpenAI Base URL.
• Enter your Novita AI API key in the API Key field.
• Replace the Base URL with: https://api.novita.ai/openai.

Step 4: Add Multiple AI Coding Models

Click “+ Add Custom Model” and add each model:

• minimax/minimax-m2
• qwen/qwen3-coder-480b-a35b-instruct
• zai-org/glm-4.6
• deepseek/deepseek-v3.1
• moonshotai/kimi-k2-0905
• openai/gpt-oss-120b
• google/gemma-3-12b-it

Step 5: Test Your Integration

• Start new chat in Ask Mode or Agent Mode
• Try multiple models across different coding scenarios
• Confirm each model returns valid responses

Troubleshooting

Issue	Possible Cause	How to Fix
Model not responding / request fails	Incorrect API key or expired key	Re-enter your Novita AI API key and ensure there are no trailing spaces. Regenerate if needed.
model_not_found or 404 error	Model name or Base URL is misconfigured	Confirm model name (e.g., minimax-m2) and set Base URL to: https://api.novita.ai/openai.
Slow or incomplete responses	Output token limit or temperature settings not optimized	Increase max_output_tokens and reduce temperature to improve stability.
Agent stops mid-task / partial edits	Agent hit safety limits or unclear instructions	Add clearer step-by-step goals to the prompt and re-run. Break large refactors into smaller actions.

Frequently Asked Questions