LobeHub
LobeHub

Changelog

Track new features, fixes, and improvements in LobeHub.

Knowledge Base: A New Way to Manage and Chat with Your Files

LobeHub Launches Knowledge Base: Intelligent File Management and Dialogue Experience

The LobeHub knowledge base is now live in both the open-source edition and the Cloud edition (LobeHub.com). This release brings file management, intelligent processing, and file-based dialogue into one workflow, so teams can organize knowledge and retrieve information with less friction.

A dedicated space for your files

We added a Files entry in the left sidebar that gives uploaded content a permanent home. Instead of juggling files across scattered entry points, you now keep everything in one stable location.

You can upload and store documents, images, audio, and video. LobeHub provides built-in online preview for common formats including PDF, Excel, Word, PPT, and TXT. The preview component is built on open-source foundations, with an architecture designed to expand to more file formats over time.

Intelligent knowledge organization

You can create unlimited knowledge bases to match different scenarios, from personal reference collections to domain-focused team libraries.

After upload, files are automatically chunked and vectorized, with fragment preview available for faster inspection. The feature also integrates the Portal interaction pattern to support quicker preview and retrieval of file content.

The current architecture reserves room for future expansion, including planned intelligent processing for multimedia files such as audio, images, and video.

Ready to use, wherever you deploy

Knowledge base features work out of the box:

  • Upload files directly inside the dialogue box with an intuitive flow
  • Clear progress feedback during uploads
  • Available in both open-source self-hosted edition and official Cloud edition

All features are open-sourced on the GitHub repository. Visit LobeHub Cloud to experience the full functionality.

v1.12.0 ~ v1.8.1

LobeHub Database Docker Image: The Final Piece of the Cloud Deployment Puzzle

LobeHub Database Docker Image Official Release

With LobeHub v1.8.0, the official database Docker image is now available. This completes the server-side deployment path we have been building, so teams can run a full cloud deployment without gaps in their setup flow.

The image ships with Postgres and NextAuth pre-configured. You get cloud data synchronization and flexible authentication out of the box, including support for third-party SSO providers like Auth0.

Lightweight but complete

The Docker image weighs only 90MB, yet provides a full database environment. We pre-configured Server Postgres and the NextAuth authentication system to ensure stable connectivity from the start. This keeps deployment simple while giving you production-grade foundations.

After deployment, cloud data synchronization is available immediately. Your data flows between client and server without additional configuration steps.

Self-hosting documentation rebuilt

To make private deployments smoother, we reorganized the deployment documentation with clearer structure:

  • Framework concepts explained up front
  • Detailed deployment case studies
  • Step-by-step self-hosting guide

Visit the official documentation to start deploying your own LobeHub service.

Improvements and fixes

  • Released official database Docker image (90MB)
  • Pre-configured Server Postgres and NextAuth authentication
  • Enabled cloud data synchronization post-deployment
  • Added Auth0 and third-party SSO provider integration
  • Reorganized self-hosting documentation with case studies

Credits

Thanks to the community for the patience while we built this part of the stack. We will keep improving the overall deployment experience.

v1.8.0 ~ v1.6.1

LobeHub v1.6: GPT-4o Mini Joins the Default Lineup

LobeHub v1.6: GPT-4o Mini Joins the Default Lineup

OpenAI's full model family has moved to GPT-4. LobeHub v1.6 follows that shift, adding GPT-4o mini to the supported models. For LobeHub Cloud users, this upgrade goes further: GPT-4o mini is now the default, replacing GPT-3.5-turbo.

The result is stronger conversations from your first message, without any configuration changes.

GPT-4o Mini: Capable and Cost-Effective

GPT-4o mini brings GPT-4-level intelligence at a smaller scale. It's fast enough for real-time interactions and capable enough for most everyday tasks—drafting, analysis, coding help, and creative work.

Use GPT-4o mini when you want:

  • Better reasoning than GPT-3.5 without the latency of full GPT-4o
  • A cost-effective default for high-volume conversations
  • Strong performance on instruction following and tool use

Switch to full GPT-4o or other providers (Claude 3.5 Sonnet, Gemini 1.5 Pro) when you need maximum capability for complex reasoning tasks.

Cloud Service: Upgraded Defaults

For LobeHub Cloud users, the service upgrade is automatic. New conversations start with GPT-4o mini by default. Existing users don't need to change any settings—the model switcher simply shows the new default first.

Cloud now supports:

  • GPT-4o mini (default)
  • GPT-4o
  • Claude 3.5 Sonnet
  • Gemini 1.5 Pro

Improvements and fixes

  • Added GPT-4o mini model configuration and parameter defaults
  • Updated LobeHub Cloud default model selection logic
  • Improved model switcher UI to highlight recommended options
  • Fixed edge cases in streaming responses for newer OpenAI models
v1.6.0 ~ v1.0.1

LobeHub 1.0: A New Foundation for Persistent, Multi-User Workspaces

LobeHub 1.0: A New Foundation for Persistent, Multi-User Workspaces

Since March, we've been rebuilding LobeHub from the ground up. Two months later, 1.0 is here. This isn't just an incremental update—it's a new architecture that enables the capabilities users have been asking for most.

The 0.x era was defined by browser storage. Fast, simple, but limited. You couldn't sync across devices, build knowledge bases, or share agents with a team. Every session started fresh. LobeHub 1.0 changes that foundation.

Server-Side Database: Data That Persists and Travels

The centerpiece of 1.0 is server-side database support. With persistent storage, conversations and agents live beyond a single browser session. Switch from laptop to desktop without losing context. Build up institutional knowledge over time instead of starting from zero.

This unlocks capabilities that were impossible in 0.x:

  • Knowledge bases: Store documents and reference them across conversations
  • Cross-device sync: Pick up exactly where you left off on any device
  • Private agent marketplaces: Share specialized agents within your team
  • Conversation history: Search and revisit past discussions

User Management: From Single-Player to Multi-Player

Alongside the database, 1.0 introduces proper user authentication and management. We've integrated both next-auth and Clerk as authentication providers, giving you flexibility based on your security needs.

The settings area has been restructured into a dedicated user panel that brings identity, preferences, and access control into one place. This is essential infrastructure for teams. Multiple users can now share a LobeHub instance with proper access boundaries and account management.

Use the new panel to:

  • Manage account settings and API keys in one place
  • Configure authentication providers (next-auth or Clerk)
  • Control workspace access for team members
  • Switch between personal and team contexts

LobeHub Cloud: Managed 1.0, Ready to Use

LobeHub Cloud is our hosted offering built on the 1.0 architecture. All the capabilities above—server-side persistence, user management, knowledge bases—are available now without any setup.

We've opened a beta waitlist with limited daily access. If you want to skip self-hosting and start using LobeHub 1.0 immediately, join the waitlist here.

Improvements and fixes

  • Migrated core storage layer from localStorage to PostgreSQL
  • Added database migration system for seamless upgrades
  • Implemented session management with secure token handling
  • Refactored settings UI into dedicated user panel
  • Added support for multiple authentication providers
  • Improved initial load performance with server-side rendering

Credits

Huge thanks to everyone who contributed to the 1.0 architecture overhaul. This release represents foundational work by the entire LobeHub team that will support the platform for years to come.

v1.0.0 ~ v0.147.0

Run Local Models Alongside Cloud AIs

Run Local Models Alongside Cloud AIs

Cloud models are powerful, but sometimes you need data to stay local. Maybe it's a sensitive project. Maybe you want to experiment without API costs. Maybe you just like the idea of owning the entire stack. LobeHub v0.127.0 now supports Ollama, giving you the same chat experience whether your model lives in the cloud or on your machine.

No separate interface to learn. No workflow fragmentation. Just point LobeHub at your local Ollama instance and start chatting.

Connect Your Local Models in One Line

Getting started is straightforward. If you already have Ollama running, connect LobeHub with a single Docker command:

*.sh
Shell
docker run -d -p 3210:3210 -e OLLAMA_PROXY_URL=http://host.docker.internal:11434/v1 lobehub/lobe-chat

That's it. LobeHub detects your local models and makes them available in the same model switcher you use for GPT-4, Claude, and others. Mix cloud and local models in the same workspace depending on what each conversation needs.

When to Use Local Models

  • Privacy-first work: Keep sensitive conversations on your machine
  • Cost control: No per-token charges for experimentation
  • Offline access: Continue working without internet connectivity
  • Model testing: Evaluate open-source models before production deployment

Improvements and fixes

  • Added automatic model discovery from Ollama endpoints
  • Fixed streaming response handling for local model compatibility
  • Improved error handling when Ollama service is unreachable

Credits

Huge thanks to the community contributor who made Ollama integration possible, and to the Ollama team for building accessible local AI infrastructure.

v0.127.0 ~ v0.125.1
LobeHub

Super-Agents on Your Team

Bring Super-Agents onto your team, starting today.