Manus 1.5 speeds up and builds full-stack apps

Manus launched version 1.5 with faster task completion, higher quality, and full-stack web app building with auth, databases and deploy. Manus-1.5-Lite is free for all users; full Manus-1.5 is for subscribers.

Manus has released version 1.5 of its autonomous agent, and the headline change is scope: the agent no longer just researches, drafts and browses on your behalf, it can now build and ship full-stack web applications, including user authentication, a database and deployment.

What changed in Manus 1.5

According to the company, the new version completes tasks faster and produces higher-quality results than earlier releases. The bigger addition is end-to-end app building. Instead of handing you a code snippet or a static mockup, Manus 1.5 aims to produce a working web app with login, persistent data and a live deployment, all from a natural-language request.

Manus is shipping two tiers. Manus-1.5-Lite is available free to all users, while the full Manus-1.5 model is reserved for paying subscribers. That split matters: it lets anyone test the workflow before committing, and it signals that the company sees the heavier model as its premium product.

Why it matters for builders

The "prompt to deployed app" category is getting crowded. Tools such as Lovable, Bolt.new and Replit already let non-engineers spin up working products, usually backed by services like Supabase for auth and data. Manus approaches the same goal from the other direction: it started as a general-purpose agent that could operate a browser and complete long tasks, and is now extending that autonomy into software delivery.

For indie hackers and small teams, this has two practical implications. First, prototyping cost keeps falling. If an agent can stand up auth, a database and hosting in one pass, validating an idea takes hours rather than days. Second, the hard part shifts to judgment: choosing what to build, reviewing generated code for security issues, and owning the result once real users arrive. Agent-built apps still need a human to check access rules, data handling and costs.

How to evaluate it

If you are already using an app builder, a fair test is to give Manus-1.5-Lite and your current tool the same spec, for example a small SaaS dashboard with sign-in and a CRUD table, and compare how much manual fixing each output needs. Pay attention to how readable the generated code is and how easy it is to move the project to your own repository and hosting.

We track Manus alongside other agent and app-building tools in the PickAI directory, so you can compare options side by side.

Source: Manus. This article is PickAI’s own summary and analysis based on the original reporting.