LobeChat review

LobeChat is built around Super-Agents that join team chats and take on work, rather than only answering one-off prompts. The homepage says agents can look for suitable tasks, pick them up without a direct assignment, and run several tasks in parallel. They can read channel history, messages, documents, and code to get context, then report completed work back in the original thread.

The workflow includes a review step: teams can inspect evidence for a delivery and accept it or send it back. LobeChat also says agents save lessons from corrections and track whether they follow those lessons later. That makes the product’s pitch less “chat with a bot” and more “delegate work in a shared conversation, then check the result.”

The pricing page lists a free plan and paid Plus, Pro, and Max tiers. Monthly billing starts at $20 for Plus; annual billing brings that to an equivalent $18 per month. Enterprise Edition is listed for private deployment, but its price is not published.

LobeChat’s clearest advantage is the team-task model. Agents can take suitable work from a conversation, handle multiple tasks at once, and return results to the thread where the request started. Teams do not have to treat every task as a separate chat, and the review-and-send-back step gives people a way to check work before accepting it.

The onboarding features matter too: according to the product description, agents can read channel history, messages, documents, and code. That gives them more context than a bare prompt, while the correction loop is intended to help agents retain lessons and track whether they apply them. The product also presents flexible connections between IM apps, agents, and models as a core feature.

There is a no-cost plan for trying the service. The pricing page lists 1 concurrent agent task, 15 MB of file storage, 100 vector entries, and 3 active Artifact sites on Free. Plus raises the task limit to 2 and includes 2 GB of storage; annual billing is $216, or $18 per month. A Reddit user who switched from OpenWebUI also praised LobeChat’s built-in web search, though that is one user’s report, not a guarantee for every setup.

The main constraint is capacity: Free allows only 1 concurrent agent task, Plus allows 2, Pro allows 5, and Max allows 10, according to the pricing page. Storage and other limits also rise by tier. Free includes 15 MB of file storage, 100 vector entries, and 3 active Artifact sites capped at 5 MB each. Plus costs $20 monthly or $18 per month with annual billing, but still caps storage at 2 GB and concurrent tasks at 2. Pro is $100 monthly ($90 annually per month); Max is $200 monthly ($180 annually per month).

The pricing page says additional usage and credit top-ups are available, but does not publish their prices. Enterprise Edition includes private-deployment options such as a self-hosted Provider, user management, and custom integration and support; its price is also not published. Teams comparing total cost should get those figures before committing.

The published Reddit discussion does not offer much direct evaluation of LobeChat’s team-agent workflow. In a thread about a governmental chatbot, commenters leaned toward LibreChat or OpenWebUI for governance or administrative control, but those are individual opinions, not a feature-by-feature assessment. Buyers with strict oversight needs should verify that LobeChat’s controls fit their requirements.

LobeChat’s Free plan costs $0, but caps you at 15 MB of file storage, 100 vector entries, one concurrent agent task and three active Artifact sites. Plus is $20 per month, or $18 per month with annual billing ($216 per year). It raises those limits to 2 GB of storage, 10,000 vector entries and two concurrent tasks.

The jump to Pro is steep: $100 monthly or $1,080 annually, for five times Plus usage, 4 GB of storage and five concurrent tasks. Max doubles the monthly price again to $200, or $2,160 per year, with 10 concurrent tasks. Annual prices are shown as monthly equivalents, so check the total yearly charge before subscribing.

Enterprise Edition has no published price. Additional usage and credit top-ups are available, but their prices are also not published. If you expect to exceed plan limits, budget for unknown add-on costs rather than assuming the listed subscription covers everything.

Moving to LobeChat means connecting your IM apps, agents and models, then deciding which chat history, messages, documents and code agents should be able to read. The product describes that team-chat workflow, but the published materials do not document migration tools, bulk imports or export formats. Treat the setup effort and portability as unconfirmed; check those details before moving important work.

Lock-in is hard to judge from the available information. LobeChat says agents save lessons from corrections, but the published material does not say whether those lessons, task histories or other data can be exported. Confirm what you can take with you before building a process around saved agent knowledge.

We can’t say what tends to make people leave based on Reddit: no Reddit posts or comments were included in the published material for this review. For a switch decision, verify connector coverage and data export with LobeChat directly rather than assuming either is supported.