Back to Home

Manus AI Glossary

Complete reference of 30+ terms related to Manus AI credits, modes, optimization, and billing. Understand every concept to make informed decisions about your AI spending.

Agent Loop

Technical

The iterative process where Manus AI analyzes context, selects tools, executes actions, and observes results until a task is complete. Each iteration consumes credits.

Checkpoint

Technical

A saved snapshot of project state in Manus web development. Checkpoints allow rollback to previous states and are required before publishing.

Context Hygiene

Optimization

The practice of removing unnecessary information from prompts to reduce token consumption without affecting output quality. A key optimization strategy that can save 10-30% on credits.

Context Window

Technical

The maximum amount of text (in tokens) that an AI model can process in a single interaction. Larger context windows allow more complex tasks but consume more credits.

Contrafactual Validation

Optimization

A technique where Credit Optimizer asks 'Would a cheaper model produce the same result?' before routing. Ensures optimization doesn't sacrifice quality.

Credit Burn Rate

Billing

The speed at which credits are consumed during task execution. Complex tasks in Max mode have higher burn rates. Monitoring burn rate helps optimize spending.

Credit Optimizer

Optimization

A middleware skill that analyzes prompt complexity and routes tasks to the cheapest model that maintains quality. Achieves 62% mean savings across 53 audited scenarios with 99.2% quality maintenance.

Credits

Billing

The unit of currency used to pay for Manus AI tasks. Each action consumes a certain number of credits based on model used and complexity. Credits are included in subscription plans or can be purchased separately.

First Principles Analysis

Optimization

A problem-solving approach used by Credit Optimizer to break down prompts into fundamental components and assess true complexity. Prevents over-routing to expensive models.

Free Tier

Billing

Manus AI's entry-level plan with limited daily credits and Standard mode only. Suitable for trying the platform before committing to a paid plan.

Function Calling

Technical

The mechanism by which Manus AI invokes specific tools or APIs during task execution. Each function call is part of the agent loop.

Max Mode

Modes

The premium processing mode in Manus AI that uses the most capable model. Required for complex reasoning, multi-step analysis, creative writing, and tasks requiring deep understanding. Costs more credits but delivers higher quality on complex tasks.

Max Plan

Billing

Manus AI's premium subscription plan at $99/month. Includes approximately 5,000 credits, unlimited Max mode access, and highest priority.

Mixed Task

Optimization

A prompt that contains both simple and complex components. For example, 'Write a quick email AND analyze this dataset.' Smart Split decomposes these into separate sub-tasks for optimal routing.

Model Selection

Optimization

The process of choosing which AI model to use for a task. Automatic model selection based on complexity is the core function of Credit Optimizer.

Overspend

Billing

When credit consumption exceeds the optimal amount for a given task, typically because a more expensive model was used when a cheaper one would suffice. Credit Optimizer prevents overspend.

Plus Plan

Billing

Manus AI's mid-tier subscription plan at $39/month. Includes approximately 1,500 credits, Standard and Max modes, and priority access.

Prompt

Core

The text instruction you send to Manus AI to initiate a task. The quality and specificity of your prompt directly affects credit consumption and output quality.

Quality Maintenance

Optimization

The percentage of output quality preserved when using optimization. Credit Optimizer maintains 99.2% quality across all scenarios, meaning virtually no degradation in results.

ROI (Return on Investment)

Billing

In the context of Manus AI, ROI measures the value of credit savings versus the cost of optimization tools. Credit Optimizer typically pays for itself within the first day of use.

Routing Strategy

Optimization

A predefined approach for handling a specific type of task. Credit Optimizer v5 includes 7 strategies: Direct Standard, Direct Max, Smart Split, Hybrid, Context Hygiene, Full Max, and Smart Testing.

Sandbox

Technical

The isolated virtual environment where Manus AI executes tasks. Each sandbox has its own file system, browser, and tools, ensuring security and preventing interference between tasks.

Skill

Core

A modular capability that extends Manus AI's functionality. Skills are reusable packages of instructions, templates, and logic that can be installed to automate specific workflows.

Smart Split

Optimization

A Credit Optimizer strategy that decomposes a complex prompt into multiple sub-tasks, routing each to the cheapest capable model. Saves 40-60% on mixed-complexity prompts.

Smart Testing

Optimization

A Credit Optimizer strategy for uncertain cases. Runs the task on Standard first, evaluates output quality, and only escalates to Max if needed. Saves credits on tasks that turn out to be simpler than expected.

Standard Mode

Modes

The default processing mode in Manus AI that uses a cost-efficient model. Suitable for routine tasks like email drafts, data lookups, simple code fixes, and formatting. Costs approximately 70% less than Max mode per task.

Task

Core

A single unit of work submitted to Manus AI. A task can range from a simple question to a complex multi-step project. Each task consumes credits based on its complexity and the mode used.

Task Routing

Optimization

The process of directing a task to the most appropriate AI model based on its complexity. Intelligent routing can save 30-75% on credits by using cheaper models for simple tasks.

Token

Technical

The smallest unit of text processed by AI models. Roughly 1 token equals 4 characters or 0.75 words in English. Token count directly affects credit consumption.

Tool Use

Technical

When Manus AI invokes external tools (browser, file system, shell, etc.) to complete a task. Tool use is part of the agent loop and contributes to credit consumption.

Ready to Optimize Your Credits?

Now that you understand the terminology, put it into practice. Credit Optimizer v5 automates intelligent routing to save you 30-75% on every prompt.