You’ve heard of AI ‘Deep Research’ tools…now Manus is launching ‘Wide Research’ that spins up 100+ agents to scour the web for you

Chinese AI startup Manus has announced a new feature called Wide Research. This innovative tool distinguishes itself from existing AI "Deep Research" tools offered by major players like OpenAI and Google. Instead of relying on a single agent for in-depth analysis, Manus’s Wide Research deploys over 100 AI agents simultaneously. This parallel processing capability is designed to tackle large-scale, high-volume tasks efficiently.

What is Wide Research?

Wide Research allows users to delegate complex tasks to hundreds of parallel AI subagents. Each subagent functions as a complete Manus instance, working concurrently to complete subtasks that contribute to a larger objective, such as analyzing product details or generating various creative assets. In a demonstration, Wide Research was used to compare 100 different sneaker models, with each subagent analyzing a single model's design, pricing, and availability. The output was a sortable matrix in both spreadsheet and webpage formats, delivered within minutes. The feature also supports creative endeavors, as shown when Manus agents simultaneously generated poster designs in 50 distinct visual styles, providing polished assets in a downloadable ZIP file. This versatility stems from Manus's system-level approach to parallel processing and inter-agent communication.

Technical Architecture and Availability

Wide Research is built on an optimized virtualization and agent architecture that boosts compute power significantly beyond Manus's previous offerings. It is automatically activated for tasks that require wide-scale analysis, eliminating the need for manual configuration. Currently, Wide Research is accessible to users on the Manus Pro plan, with a gradual rollout planned for Plus and Basic plan subscribers. Manus subscription tiers are structured as follows:

• Free ($0/month): Offers 300 daily refresh credits, Chat mode, and 1 concurrent/scheduled task.
• Basic ($19/month): Includes 1,900 monthly credits (+1,900 bonus limited offer), 2 concurrent/scheduled tasks, advanced models in Agent mode, media generation, and access to exclusive data sources.
• Plus ($39/month): Provides 3 concurrent/scheduled tasks, 3,900 monthly credits (+3,900 bonus), and all Basic features.
• Pro ($199/month): Grants 10 concurrent/scheduled tasks, 19,900 credits (+19,900 bonus), early beta access, a Manus T-shirt, and the full feature set.
An annual payment option offers a 17% discount. Each Manus session operates on a dedicated virtual machine, providing users with orchestrated cloud compute managed through natural language commands. The Wide Research architecture avoids predefined agent roles; instead, every subagent is a general-purpose instance capable of any task, ensuring flexible and scalable workflows.

Benefits and Challenges Compared to Deep Research

The core promise of Wide Research is its parallelism, which aims to deliver faster results and a greater diversity of outputs compared to single-agent Deep Research models. However, Manus has yet to release specific benchmarks or detailed technical comparisons to substantiate these claims. Information regarding resource efficiency, coordination mechanisms among agents, or validated accuracy improvements is also pending.

Broader Context and Limitations

Multi-agent systems have demonstrated varied success in the AI community. For instance, some users of Anthropic Claude’s subagents have reported issues such as slow performance, high token consumption, and poor coordination. Manus acknowledges that Wide Research is an experimental feature and may encounter similar challenges as its development progresses.

Looking Ahead

Manus's Wide Research represents a significant advancement in the pursuit of scalable, multi-threaded AI collaboration. If it successfully addresses the coordination and efficiency hurdles observed in other multi-agent systems, its general-purpose subagent design could fundamentally alter AI workflows. The industry will be keenly observing its development and practical impact.

---

Frequently Asked Questions (FAQ)

Wide Research Functionality

Q: What is Manus's Wide Research? A: Wide Research is a new feature from Manus that utilizes over 100 AI agents simultaneously to tackle large-scale, high-volume research tasks, differing from single-agent "Deep Research" tools. Q: How does Wide Research differ from other AI research tools? A: Unlike tools that use a single AI agent, Wide Research deploys a multitude of agents in parallel to process tasks more broadly and concurrently. Q: What types of tasks can Wide Research handle? A: Wide Research is designed for complex, large-scale tasks, including data analysis (like comparing product features across many items) and creative asset generation (like producing designs in multiple styles). Q: Is Wide Research available to all Manus users? A: Currently, Wide Research is available to Manus Pro plan users. Rollout to Plus and Basic plan subscribers is planned. Q: What are the benefits of using multiple AI agents in Wide Research? A: The benefit lies in increased speed and the ability to handle tasks that require processing a large volume of information or generating diverse outputs simultaneously.

Technical Aspects and Availability

Q: How is Wide Research technically implemented? A: It leverages an optimized virtualization and agent architecture that scales compute power significantly. Each subagent acts as a fully capable Manus instance. Q: Does Wide Research require manual setup for each task? A: No, Wide Research activates automatically for tasks that require wide-scale analysis without manual configuration. Q: What are the different Manus subscription plans and their features? A: Manus offers Free ($0/month), Basic ($19/month), Plus ($39/month), and Pro ($199/month) plans, each with varying credits, concurrent tasks, and feature access.

Performance and Limitations

Q: Has Manus provided benchmarks for Wide Research? A: No, Manus has not yet released specific benchmarks or detailed technical comparisons to validate the performance claims of Wide Research against Deep Research models. Q: Are there any known limitations or challenges with Wide Research? A: Manus acknowledges that Wide Research is experimental and may face challenges similar to other multi-agent systems, such as potential issues with performance, resource efficiency, and agent coordination.

Crypto Market AI's Take

Manus's introduction of "Wide Research" marks a fascinating development in the AI agent landscape, particularly in how it approaches large-scale data processing. The concept of leveraging a hundred concurrent agents to tackle tasks like comprehensive product comparisons or generating diverse creative assets simultaneously offers a glimpse into a future where AI can handle complex, distributed workloads with unprecedented efficiency. This approach could have significant implications for industries requiring rapid analysis of vast datasets, including market research in the cryptocurrency sector. At Crypto Market AI, we explore how AI agents are transforming various aspects of finance, from predicting market trends to automating trading strategies. The ability to scale research capabilities with parallel agents, as demonstrated by Manus, aligns with the broader trend of AI agents becoming more sophisticated and integral to business operations.

More to Read:

• AI Agents: Capabilities, Risks, and Growing Role
• AI-Driven Crypto Trading Tools Reshape Market Strategies in 2025
• Understanding AI Crypto Scams

Source: Originally published at VentureBeat on July 31, 2025.