Guide · Beta
Bring your own key
Run Sonny's AI on your own OpenAI or Anthropic account — or a self-hosted, OpenAI-compatible server — with unlimited AI conversations. A one-time $249 unlock, no extra subscription. Here's what you get and how to set it up.
Want the full pitch and cost comparison? See the Bring Your Own Key overview.
What the unlock gets you
- Unlimited AI conversations
- Monthly conversation allowances disappear. Your provider bills you for the tokens your selected model uses.
- Your provider, your model
- Connect OpenAI, Anthropic (Claude), or any OpenAI-compatible endpoint — including a self-hosted Ollama or vLLM server. Choose from curated models or enter any model ID.
- Control where AI processing happens
- Every customer-facing response is generated on your account, under your provider agreement. Ideal for custom server environments and data-residency requirements.
- Keys stay secret
- Your API key is encrypted at rest and never displayed again — only a hint like "sk-a…4f2a" so you know which key is connected.
- All AI features included
- The auto-responder, AI reply drafts, and polish all run through your key. Behavior, confidence thresholds, and office-hours schedules work exactly as before.
- Knowledge base stays managed
- Your knowledge sources stay managed by Sonny at no extra cost. Your connected provider generates the customer-facing answers.
Setting it up
- 01
Unlock Bring your own key
Open AI Copilot and find the AI engine card. Under Bring your own key, select Unlock for $249. It is one purchase per workspace, and only the workspace owner can buy it.
- 02
Set up your key
Select Set up your key to open Connect your own AI. Under Provider, choose OpenAI, Anthropic (Claude), or OpenAI-compatible. Enter the API key and choose a Model. OpenAI-compatible servers also need their final public Endpoint URL over HTTPS; include the
/v1path when the server uses one. Add Basic Auth or custom access headers if your proxy requires them. - 03
Test connection
Select Test connection. Connectivity checks that Sonny can reach and authenticate with the provider. Tool calling checks that the model can search knowledge and hand off to a teammate. Both must pass before the connection can be enabled.
- 04
Save & enable
Select Save & enable. The AI Copilot page now shows Running on your key. Use Manage connection to change the provider, credentials, or model later.
A note on cost
Your exact cost depends on your provider, model, and token usage. Check your provider's current pricing, then use its usage dashboard to track what your conversations actually cost.
Good to know
- Does my self-hosted endpoint have to use HTTPS?
- Yes. Sonny accepts only public https:// URLs for self-hosted providers. If you previously connected an http:// endpoint, that connection has been switched off and you'll see a note in AI Copilot settings: put your server behind an HTTPS proxy, then re-enter the URL to turn it back on.
- Can I protect a public Ollama server?
- Yes. Ollama's local API does not authenticate requests itself, so put it behind an HTTPS reverse proxy and enter that endpoint's HTTP Basic Auth credentials in Sonny. You can also add up to 10 write-only custom headers for gateways such as Cloudflare Access. When your proxy handles authentication you can leave the API key blank — Basic Auth replaces Bearer API-key authentication, so a placeholder key would never be sent. If your gateway needs both, configure its API key in a different custom header. Credentials and header values are encrypted and never returned to your browser.
- What happens if my key stops working?
- Customers are never left hanging. If your provider rejects a request — an expired key, a rate limit, a server outage — the auto-responder immediately routes those conversations to your team, and workspace owners get a notification with the error. Fix the key (or toggle back to Sonny AI) and everything resumes.
- Can I switch back to Sonny AI?
- Anytime. The card has an on/off toggle — turn it off and Sonny AI takes over instantly (with an active AI plan). Your key and settings are kept so you can switch back with one click. You can even keep an AI plan and BYOK active together; whichever is enabled handles responses.
- Which models should I use?
- Use a fast, tool-capable model such as a Claude Haiku-class or OpenAI mini-class model. Smaller self-hosted models can work, but expect lower answer quality and less reliable escalation; the built-in tool-calling test will tell you if a model isn't up to the job.
- Do knowledge base pages still have a cap?
- Yes — BYOK includes a 5,000-page knowledge cap because Sonny continues to manage your knowledge sources separately from provider usage. If you also have an AI plan with a higher cap, you keep the higher one.
- Is my API key safe?
- Your credentials are encrypted at rest, never shown again after saving, and used only when Sonny sends a request to your configured provider. Self-hosted endpoint URLs are restricted to the public HTTPS endpoint you configure.