LobeHub (formerly LobeChat) vs Jan vs WeKnora

Side by side, as alternatives to ChatGPT Plus. Rows that differ are highlighted; the strongest value in a row is marked. Unknown means no source yet.

LobeHub (formerly LobeChat)lobehub/lobehub

Janjanhq/jan

WeKnoraTencent/WeKnora

ChatGPT Pluspaid product

Cost and coverage

Price

$0 software; you run it

Plus $20/month; Pro from $100/month source ↗

Chat assistant

Partly

Yes

Chat with your documents

Partly

Yes

Web search answers

Yes

Jobs covered

1 of 3 jobs

All 3

Popularity and activity (from GitHub)

Stars

83,366Top

45,023

34,870

—

Gain this week

Unknown

—

Signal

Building history

—

Commits / 4 wk

1,103Top

195

274

—

Latest release

v2.2.20 · 2026-10-18

v0.8.7 · 2026-10-20Top

v0.8.3 · 2026-10-14

—

Commits / 12 wk

3,046

748

976

—

People committing / 12 mo

129+

46+

99+

—

Issues opened / closed, 30 days

103 / 52

119 / 137

353 / 323

—

Facts, with sources on each project page

Runs on my machine?Differs

Yes, self-hosted via Docker; desktop app also available

Yes, desktop app running models offline

Yes, self-hosted via Docker Compose, Kubernetes or a single Lite binary

Hosted by the vendor

Code licenceDiffers

LobeHub Community License (Apache-2.0 + conditions); not OSI-approved

Apache-2.0 (attribution requested)

MIT (third-party components keep their own licences; GitHub shows NOASSERTION)

—

Model weights licenceDiffers

No bundled weights; licence depends on the model or API you connect

No bundled weights; licence depends on the model you download

No bundled chat weights; licence depends on the model you connect

—

Commercial useDiffers

Unmodified commercial use OK; distributing derivatives needs a paid licence

Yes, Apache-2.0 app; check the licence of the model you use

Yes, MIT code; check third-party component and model licences

—

Paid services needed?Differs

Self-host guide requires an OpenAI key; local Ollama documented as alternative

None required; cloud model providers optional

None required with local Ollama; paid model APIs, web search APIs and WeKnora Cloud optional

Subscription

HardwareDiffers

Docker Compose self-hosting: minimum 2 CPU cores, 4 GB RAM, 20 GB storage; production 4+ cores, 8+ GB RAM, 50+ GB storage

macOS 8GB RAM for 3B, 16GB for 7B, 32GB for 13B models; GPU optional

Standard Docker deployment: 4-core CPU and 8 GB RAM to start, excluding Ollama model weights; x86 CPUs need x86-64-v2; disk scales with knowledge base

—

Operating systemsDiffers

Docker self-host; desktop app on Windows, macOS, some Linux distributions

Windows 10+, macOS 13.6+, Linux (deb, AppImage, Flathub)

Official installation docs describe Docker/Kubernetes servers and a source-built macOS desktop app. Windows/Linux/macOS desktop build targets exist, but the docs warn that release installers are not published

—

Setup effortDiffers

Medium; setup script + docker compose, or one-click cloud deploy

Easy; download installer (exe, dmg, deb, AppImage) or app stores

Medium; Docker Compose with several services, then configure models in the web UI

—

Stars over time

Switch to Growth % to compare projects of different sizes fairly.

LobeHub (formerly LobeChat)
Jan
WeKnora

Focus a point, use arrow keys, or tap for its value.

Tracked daily since 15 Oct 2026. Dashed history: estimated from public records of when people starred the repo. Lines connect observations only; gaps contain no measured points.

Growth % starts at each project's first observed point in the shown range. Zero-star starts are unavailable.

Notes for a video or post

A plain summary with today's date and a link to the sources. Check each project page before you publish.

LobeHub (formerly LobeChat) vs Jan vs WeKnora (as alternatives to ChatGPT Plus)
Figures from RepoQuarry, 21 Oct 2026. Sources: https://repoquarry.dev/compare?ids=lobehub,jan,weknora&product=chatgpt-plus

ChatGPT Plus: Plus $20/month; Pro from $100/month
LobeHub (formerly LobeChat): free software; does 1 of 3 ChatGPT Plus jobs
Jan: free software; does 1 of 3 ChatGPT Plus jobs
WeKnora: free software; does 1 of 3 ChatGPT Plus jobs

- LobeHub (formerly LobeChat): 83,366 GitHub stars, 3,046 commits in 12 weeks, licence LobeHub Community License (Apache-2.0 + conditions); not OSI-approved
- Jan: 45,023 GitHub stars, 748 commits in 12 weeks, licence Apache-2.0 (attribution requested)
- WeKnora: 34,870 GitHub stars, 976 commits in 12 weeks, licence MIT (third-party components keep their own licences; GitHub shows NOASSERTION)

Where they differ:
- Runs on my machine?: LobeHub (formerly LobeChat) Yes, self-hosted via Docker; desktop app also available; Jan Yes, desktop app running models offline; WeKnora Yes, self-hosted via Docker Compose, Kubernetes or a single Lite binary
- Code licence: LobeHub (formerly LobeChat) LobeHub Community License (Apache-2.0 + conditions); not OSI-approved; Jan Apache-2.0 (attribution requested); WeKnora MIT (third-party components keep their own licences; GitHub shows NOASSERTION)
- Model weights licence: LobeHub (formerly LobeChat) No bundled weights; licence depends on the model or API you connect; Jan No bundled weights; licence depends on the model you download; WeKnora No bundled chat weights; licence depends on the model you connect
- Commercial use: LobeHub (formerly LobeChat) Unmodified commercial use OK; distributing derivatives needs a paid licence; Jan Yes, Apache-2.0 app; check the licence of the model you use; WeKnora Yes, MIT code; check third-party component and model licences
- Paid services needed?: LobeHub (formerly LobeChat) Self-host guide requires an OpenAI key; local Ollama documented as alternative; Jan None required; cloud model providers optional; WeKnora None required with local Ollama; paid model APIs, web search APIs and WeKnora Cloud optional
- Hardware: LobeHub (formerly LobeChat) Docker Compose self-hosting: minimum 2 CPU cores, 4 GB RAM, 20 GB storage; production 4+ cores, 8+ GB RAM, 50+ GB storage; Jan macOS 8GB RAM for 3B, 16GB for 7B, 32GB for 13B models; GPU optional; WeKnora Standard Docker deployment: 4-core CPU and 8 GB RAM to start, excluding Ollama model weights; x86 CPUs need x86-64-v2; disk scales with knowledge base
- Operating systems: LobeHub (formerly LobeChat) Docker self-host; desktop app on Windows, macOS, some Linux distributions; Jan Windows 10+, macOS 13.6+, Linux (deb, AppImage, Flathub); WeKnora Official installation docs describe Docker/Kubernetes servers and a source-built macOS desktop app. Windows/Linux/macOS desktop build targets exist, but the docs warn that release installers are not published
- Setup effort: LobeHub (formerly LobeChat) Medium; setup script + docker compose, or one-click cloud deploy; Jan Easy; download installer (exe, dmg, deb, AppImage) or app stores; WeKnora Medium; Docker Compose with several services, then configure models in the web UI

Stars show popularity, not quality. Check each project page before publishing.

Want share images and a ready video description? Open a project's creator kit.

Numbers: GitHub, fetched nightly. Facts: read from each project's own documentation, with the source on its page. How we check. lobehub/lobehub ↗janhq/jan ↗Tencent/WeKnora ↗

LobeHub (formerly LobeChat) vs Jan vs WeKnora

Stars over time

Notes for a video or post

Cookie preferences

Report a problem