Update from code changes: bring your own model settings - #7134
Update from code changes: bring your own model settings#7134mintlify[bot] wants to merge 10 commits into
Conversation
|
Preview deployment for your docs. Learn more about Mintlify Previews.
|
|
Preview deployment for your docs. Learn more about Mintlify Previews.
💡 Tip: Enable Workflows to automatically generate PRs for you. |
|
Update from code changes updated this PR. Clarified BYOK as contracted Enterprise add-on hidden without entitlement Source PRs
|
- Updated quickstart.mdx Mintlify-Source: dashboard-editor
- Updated quickstart.mdx Mintlify-Source: dashboard-editor
- Updated quickstart.mdx Mintlify-Source: dashboard-editor
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 4fe8dfb. Configure here.
|
Update from code changes updated this PR. Added OpenRouter default and provider compatibility notes; reverted broken quickstart edits Source PRs
|
|
Closing as superseded. This PR opened Aug 26; the dedicated Bring your own model page landed on main Aug 28 in #7174 (style pass in #7177). It's in the nav and linked from The two details here that weren't already on that page — the HTTPS/publicly-reachable base URL constraint, and the AI model section being hidden without the add-on — are ported in #7197, along with a cross-link from Note for the next sync: BYOK config isn't assistant-specific (it covers the agent and automations too), so it belongs on |
…link (#7197) Ports two details from the superseded bot PR #7134 into the dedicated bring your own model page: - The OpenAI-compatible base URL must use HTTPS and point to a publicly reachable host. - The AI model section only appears in the dashboard when the contract includes the add-on, so readers who can't find it know why. Also adds a cross-link from the assistant configuration page, which had no path to the bring your own model page. Placed in Manage billing next to the existing credits link rather than as its own section, since the configuration covers the agent and automations too, not just the assistant. Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
1 participant
Summary
Documents the new self-serve bring-your-own-key (BYOK) settings that let Enterprise deployments run Mintlify AI products on their own LLM provider and API key.
Changes
Context
Enterprise customers can now configure the LLM that powers Mintlify AI products from the dashboard under Settings → API keys. The AI model section supports OpenRouter (default, full compatibility), Anthropic, and OpenAI-compatible providers with a curated model list per provider. API keys are stored securely, never shown after saving, and must be re-entered when provider settings change. The section is hidden entirely for deployments without the BYOK add-on.
https://github.com/mintlify/mint/pull/10437
Update history
Note
Low Risk
Documentation-only updates with no runtime, auth, or data-handling code changes.
Overview
Adds a Bring your own model section to the assistant configuration docs (English plus Spanish, French, and Chinese), placed after bot protection and before billing.
The new content documents the Enterprise add-on requirement, dashboard flow on API keys (AI model → Bring your own model), supported providers (OpenRouter as default with full AI product support; Anthropic and OpenAI-compatible with limited compatibility), HTTPS base URL rules for OpenAI-compatible endpoints, secure API key storage/reuse/rotation, and disabling the toggle to remove BYOK config. Localized pages also gain BYOK-related keywords and anchor ids (
bring-your-own-model).Reviewed by Cursor Bugbot for commit 63b3565. Bugbot is set up for automated code reviews on this repo. Configure here.