Documentation and answers
Build product manuals, SOPs and public or private help centers. Import existing material, translate content and help readers find answers with AI.
Docsie Self-Hosted · Kubernetes preview
Your knowledge.
Your infrastructure.
Turn your existing knowledge into documentation, searchable help centers, training and AI-assisted answers. Run Docsie on your own Kubernetes cluster and choose local or hosted AI services for each workflow.
Free for personal use. Free 30-day business evaluation for up to 100 users. Docsie is proprietary software. Helm charts, installation guides and agent setup instructions are available on GitHub.
Put your knowledge to work
Training and paid courses
Turn videos and documents into courses, quizzes and certifications. Connect commerce services to sell learner access, and reuse approved content in PowerPoint training decks.
Comparison and review
Compare manuals, specifications, websites and recordings. Review differences, gaps and conflicting information with source references, then improve your documentation.
Choose the workflows your team needs. The use-case guide explains their dependencies and verification steps for self-hosting.
Install in three steps
Pull the public images
Start with the release image list for your chosen services and CPU architecture. Public Dokuta images are available for AMD64 and ARM64; check the release guide for access to the remaining application components.
Image list on GitHub →Helm on your cluster
Clone the repository, build chart dependencies, set your hostname, storage and admin account in a values file, then run the explicit-kubeconfig installer. PostgreSQL, Redis, MinIO and App Search are included in the platform chart.
Kubernetes install guide →Point it at your AI
Connect Ollama or any OpenAI-compatible endpoint for generation. Choose local, hosted or mixed providers independently for text, vision, embeddings, transcription and speech.
Ollama walkthrough →# v0.2.0-preview.5
git clone https://github.com/LikaloLLC/docsie-self-hosted.git
cd docsie-self-hosted
git checkout v0.2.0-preview.5
helm dependency build charts/docsie-platform --skip-refresh
bash scripts/install-kubernetes.sh /path/to/kubeconfig docsie /path/to/values.yaml
git clone https://github.com/LikaloLLC/docsie-self-hosted.git
cd docsie-self-hosted
git checkout v0.2.0-preview.5
helm dependency build charts/docsie-platform --skip-refresh
bash scripts/install-kubernetes.sh /path/to/kubeconfig docsie /path/to/values.yaml
Installing with an AI coding agent? Point it at AGENTS.md and describe the outcome you want, such as "internal SOPs plus video-to-docs on our existing cluster using our Ollama server". The agent selects the required components, prepares configuration and verifies the result.
What you can install today
| Feature | Availability | Requirements |
|---|---|---|
| Core knowledge base | Included | Kubernetes cluster. The platform chart brings PostgreSQL, Redis, MinIO, Elasticsearch, web and worker processes, document and image converters. |
| Text, vision and embeddings | Configurable | Ollama or any OpenAI-compatible endpoint, local or hosted, for chat, retrieval, vision and embeddings. |
| Dokuta video-to-docs | Included | Text and vision models, plus a transcription endpoint for narrated recordings. |
| Transcription | Bring your own | Any Whisper-compatible endpoint on your network or a hosted provider. |
| Speech generation (Chatterbox) | Optional | Enable the Chatterbox install for CPU or NVIDIA, or the native Mac CPU installer. Off by default, so a basic install carries no extra model overhead. |
| Live voice agents | Optional | Add LiveKit and the voice-agent worker, then connect speech recognition, a language model and text-to-speech. |
| Learn, Forms, translation, comparison, policy review, automation | Included | Application features; each uses the AI endpoints configured above. |
| AWS one-click and offline bundle | Not yet | Kubernetes only today. Offline documentation-reader delivery is a separate workflow; see the deployment guide. |
Read the release status before choosing a deployment. It states exactly what was validated.
Simple access, clear terms
Free for your own projects
One active user, personal non-commercial use. Bring your own machine or cluster and your own AI access. Work for an employer, client or business requires a business license.
30 days. Up to 100 users.
Evaluate Docsie in your environment with your team. Choose the workflows you want to test, connect your models and invite your team. We can help with installation and confirm your evaluation terms.
Keep your deployment
Continue with a paid license after your evaluation. Continue using the same installation and content with a commercial license. Infrastructure, storage and inference costs remain yours.
Common questions
Is Docsie open source?
Docsie is proprietary software with public Helm charts, installation scripts and documentation. Personal, non-commercial use is free for one active application user. Businesses can evaluate it for 30 days with up to 100 users, then continue with a commercial license.
Where do I get the container images?
The image guide lists public images, supported architectures and pull commands. Check the selected release for the complete set of components and any additional image access your installation needs.
What infrastructure do we need?
A Kubernetes cluster, persistent storage, Helm 3 and kubectl. The chart includes the database, cache, object storage and App Search. Start planning a small evaluation with 16 GiB of RAM for the application and search; model memory is additional. Workload and concurrency affect sizing. See the sizing guide for estimates and larger deployments.
Can we use our own AI models?
Yes. Connect Ollama or another compatible model server for text and vision, configure local embeddings and transcription, and add Chatterbox for speech. You can also mix local and hosted providers. For a private-only setup, configure each AI service and its fallbacks to use your network; review model downloads and external integrations separately.
Can I run Docsie on a Mac or ARM server?
Docsie runs in Kubernetes, including local clusters where the selected images support your hardware. Dokuta images support Linux AMD64 and ARM64; check every required component before choosing an ARM deployment. Mac users can also run the optional Chatterbox speech service natively on CPU. That installer installs speech generation, not the full Docsie application.
What happens after the business trial?
Continue using your existing installation and content with a paid business license. Contact us before your evaluation ends to arrange the license for your team.
Try Docsie on your infrastructure
Start with the installation guide, or talk to us about your deployment. Tell us what you want to build, where your data needs to stay and which models you want to use.
Related: Self-hosted documentation platform · On-premise knowledge base · Air-gapped documentation · Self-hosted ChatGPT for docs