Integrations / Ollama
DeployDocumented
Run Ollama on your computer or in the cloud
One command starts Ollama on this computer for free, or on a Badgr cloud GPU when yours is too small.
The steps follow the tool's documented settings. No tested run is recorded for this page yet.
Why use Badgr for Ollama
- Badgr interface
- Serve
- Where Ollama fits
- Model server
- Why Badgr
- Cloud and private model hosting
- The same command runs Ollama on your own computer for free, or on a cloud GPU when yours is too small.
- Cloud runs show the price first and ask you to confirm, with a cost cap.
- Badgr starts a pinned Ollama version, so an update does not change it under you.
What you can do
- Chat with an open model on your own machine
- Use a larger model than your GPU can hold, in the cloud
- Point Open WebUI or your code at the Ollama address
Quickstart: Local or cloud
badgr launch ollama --target local --model qwen3:8b badgr launch ollama --target cloud --model qwen3:8b --max-cost 5
How it works
- 1.Badgr checks Docker and your GPU
- 2.It starts the pinned Ollama version here, or on cloud if you chose it
- 3.It pulls the model you asked for
- 4.It prints the address to use
What you get
- An Ollama address your tools can call
- Local: models kept in saved storage. Cloud: files are removed when you stop it.
Requirements
- The Badgr CLI installed and logged in (npm install -g badgr-cli)
- Docker running (for local)
- For cloud: a Badgr account with a funded balance
What it costs
- Local runs are free and need no Badgr account.
- Cloud runs bill by use. Badgr shows the rate and asks you to confirm before anything starts.
- Pass --max-cost to cap a cloud run.
See pricing for how billing works.
Proof
No tested run is recorded for this integration yet. A receipt and example output will be added here after a real run, not before.
Troubleshooting
Model gone after stopping on cloud
Cloud files are removed on stop, so the model downloads again next time. Local runs keep models in saved storage.
Docker not running
Start Docker, then rerun the command. Badgr checks Docker before it starts.
GPU too small locally
Badgr explains why and offers the cloud command with the price first.
Developer reference
Ready to start with Ollama?
Every run or service has a spending limit you set before it starts.