## Description The LSP's `aiChatAvailableModels` GraphQL query returns `Internal server error` on SM instances with an offline cloud license and self-hosted models configured. This **does not** affect functionality of agentic chat– in fact, it is expected as this query fetches available models for user-level model selection, which is currently not supported with self-hosted models \[[ref](https://gitlab.com/gitlab-org/gitlab/-/work_items/594381#note_3239316074)\]. {width="338" height="137"} However, this shouldn’t surface as an error that can confuse users. ## Steps to Reproduce **Prerequisites**: - Create a project in the testing instance and clone locally. - Ensure you have the GitLab Workflow Extension installed on VSCode. - Setup GitLab authentication with your instance: - Open the command palette (Cmd+Shift+P) \> type in `GitLab: Authenticate` - `Manually enter instance URL` \> Enter your instance URL - Select `Create a token` - Ensure the account you just authenticated with is selected. **Steps**: - Open the cloned project and open agentic chat in the Activity Bar. - Observe error message. ## Acceptance Criteria - [ ] Update resolver to gracefully handle offline license with self-hosted models case.