Key points at a glance
- Ollama is a local model runtime; Enclave is an assistant world built on top of models—they operate at different layers.
- Enclave can use Ollama as a local model backend, getting both local privacy and a complete experience.
- With Ollama alone you get a model interface; with Enclave you get memory / proactivity / multiple experts / deliverables.
Side-by-side comparison
| Dimension | Enclave | Ollama |
|---|---|---|
| The layer it addresses | An assistant world on top of models | A local model runtime |
| Long-term memory | Built-in structured long-term memory | Not applicable (provides inference only) |
| Proactivity | Experts proactively remind and follow up | Not applicable |
| Multiple characters and deliverables | Multiple experts, group chats, can produce PPT / Word / Excel | Only a model inference interface |
| How the two relate | Can be plugged in as a local model backend | Can serve as one of Enclave's model sources |
| Privacy | Self-hosted + local models can be fully offline | Runs locally, offline by nature |
How to choose
If you just want to run a model locally and call the API yourself, Ollama is enough; if you want a complete assistant world built on top of a local model—one that remembers you, proactively helps you, and can produce documents—use Enclave and plug Ollama in as the local backend. The two don't conflict; in fact, they're a perfect match.
FAQ
Which large language models does Enclave support?
Enclave's model layer is fully swappable: OpenAI, Anthropic, Google, DeepSeek, and local Ollama / vLLM can all be configured. Different characters can even use different models, freely allocated by scenario and cost.What do I need to self-host Enclave? Do I need a GPU?
Not necessarily a GPU. If you use cloud model APIs (OpenAI, Anthropic, etc.), an ordinary server or home machine that can run Docker is enough — inference happens in the cloud, and your machine only runs the app and database. You only need the corresponding VRAM and compute if you want offline inference with local models (Ollama / vLLM).Is my conversation data safe? Will it be used for training?
In self-hosted mode, your conversations live only on your own drive — they never leave your machine and are never used by any third party for training. The model layer can connect to cloud APIs like OpenAI and Anthropic, or be swapped for local Ollama / vLLM to run fully offline; you decide which kind of conversation goes to which model.
Ready to try it?
Open it in your browser — no credit card, no install.