
Argent for Ollama
Argent fronts your Ollama server: people and apps call Argent, and Ollama answers only Argent. Here is how Argent fronts it, how it connects and what your team sets.
What Argent does with Ollama
Argent fronts your Ollama server. People and apps call your Argent route; Argent checks the request and forwards it to the model server, which answers only Argent.
- A named agent with a human owner. Every caller uses its own Argent agent key, tied to an accountable person. Identity tethering
- Only the models you allow. Policy decides which agents may use which local models. Model control
- Sensitive data handled even on a local model. Prompts and replies are scanned and redacted or blocked as your policy says. Sensitive data redaction
- Every call signed. Each decision is a hash-chained, signed audit record, with the session ready to replay. Session replay
How Ollama connects
Ollama reaches Argent in one way.
- Gateway base URL. Callers point their OpenAI-compatible base URL at your Argent route with their Argent agent key. Argent registers the local model server as an upstream and forwards each checked request to it.
Proven with the real tool. Argent's team has driven Ollama itself through a live Argent gateway and checked the signed records it produced.
The same steps are in the Argent console under Integrations > AI Clients, and each agent's Connect tab gives the copy-paste values for your tenant.
What you configure
What your team sets in Ollama. These are the settings the Argent console shows for this tool. Values in angle brackets come from your tenant.
- Register Ollama as an upstream. In the Argent console add an on-prem upstream with the Ollama URL (http://<host>:11434). Clients then call https://<ARGENT_GATEWAY>/<tenant>/<ollama-route>/v1 with an Argent agent key (OpenAI-compatible API).
- Lock Ollama to Argent. Keep OLLAMA_HOST at the default 127.0.0.1:11434 on the Ollama host, or bind it to a private interface and allow only the Argent gateway IP at the host firewall.
- Model pulls and cloud. Set HTTPS_PROXY (not HTTP_PROXY) for the Ollama server to pull models; install the proxy CA as a system certificate. OLLAMA_NO_CLOUD=1 disables Ollama Cloud models.
Make Argent the only path
Make Argent the only path for Ollama. These controls are documented by the platform each one belongs to (linked); pair them with the Argent settings above.
- systemd drop-in (Linux). Binds Ollama locally and disables cloud models. Vendor docs
- Windows / macOS service environment. Same variables for desktop installs. Vendor docs
- Host firewall: inbound 11434 only from Argent. Other machines cannot call Ollama directly. Vendor docs
Same settings everywhere it runs. The Argent fleet setup script and certificate bundle give every service, container and build runner the same proxy and certificate settings.
Want to see which AI tools people already run before you push anything? Start with Shadow AI discovery in watch mode. Watch before you enforce
Check that it works
Send one request and look for the signed row. Send one request through Ollama, then open the Argent console live stream and find the signed model-call row under the agent's name.
Every request also appears on the live stream with the agent's name and owner, and in the session view. Session replay
Where to go next
Keep going. Pages that cover the same ground from another angle.
- Use case: Rolling out copilots and chat safely
- Use case: AI in regulated and air-gapped environments
- Deployment: Air-gapped and on-prem
- Goals: See all AI in your company, Data loss prevention, Budget and token spend, Model control and routing, Agent identity and ownership
- Capabilities: Keyless agents, Prompt injection defense, Response screening, One-click audit evidence
- Related integrations: Warp, Open WebUI, LibreChat
- All integrations: Browse every AI client, provider, vault and identity source
Want this running on your own use cases?
Book a demo