mirror of
https://github.com/tiennm99/composes.git
synced 2026-10-15 10:14:47 +00:00
tiennm99
23b9bf8bf6
Drop the history and roadmap asides: which services predate the collection's
conventions, the unwired Open Web UI plan, the generic clone-and-troubleshoot
boilerplate. Services that publish ports or set `restart:` now simply say so.
couchbase, openvpn-as and traffmonetizer had two-line READMEs; give them the
ports, variables and storage the root README promises. traffmonetizer reads
${TOKEN} and had no .env.example, so add one.
32 lines
779 B
Markdown
32 lines
779 B
Markdown
# ollama
|
|
|
|
[Ollama](https://ollama.com) LLM server. CPU-only.
|
|
|
|
Uses `docker-compose.yml`, publishes `11434`, and sets
|
|
`restart: unless-stopped` and `container_name: ollama` — unlike the
|
|
platform-managed services described in the [root README](../README.md).
|
|
|
|
## Usage
|
|
|
|
```sh
|
|
docker compose exec ollama ollama pull llama3.2
|
|
docker compose exec ollama ollama run llama3.2
|
|
```
|
|
|
|
The API is on `http://localhost:11434`:
|
|
|
|
```sh
|
|
curl http://localhost:11434/api/generate -d '{
|
|
"model": "llama3.2",
|
|
"prompt": "Why is the sky blue?",
|
|
"stream": false
|
|
}'
|
|
```
|
|
|
|
For GPU, add `deploy.resources.reservations.devices` for the NVIDIA runtime —
|
|
see the [Ollama Docker docs](https://hub.docker.com/r/ollama/ollama).
|
|
|
|
## Storage
|
|
|
|
`ollama` at `/root/.ollama` — pulled models and manifests.
|