Skip to main content

Improving Usability with OpenRouter Model Discovery

Feature Request

We should update the OpenRouter integration so that, instead of defaulting to a chat completion call to check the API, we use the list models call. This will allow us to retrieve all available model information directly.

With this change, users can easily select from the available free models without needing to configure a separate custom model for free mode. This will significantly improve overall app usability.

Since OpenRouter follows the OpenAI specification, we can make the feature more robust by:

  • Providing the LLM gateway (OpenAI-compatible) URL and API key.

  • Retrieving the model list.

  • Allowing users to select their desired model from that list.

    This useful for other LLM gateway provider like Vercel.

Status: Completed5 comments

Log in to comment and vote

Comments5

  • Ngoc Nguyen changed status to Completed
    Team•

    May 11

    Pinned

    we have added official support for OpenRouter, check here for more details:

    https://feedback.typingmind.com/changelog/official-support-for-more-ai-providers

    • Aquamarine Camel

      •

      May 13

      Awesome! Thank you!

  • Amethyst Bagel

    •

    Apr 20

    •

    Merged request

    •

    1 vote

    🔄 need ability to refresh OpenRouter latest models (custom model)

    at the moment “custom model” using OpenRouter only allows to select the models that were available when the custom model was initially set up.

    As new models rollout, model list is stale, outdated, not showing newer models released since then. This forces us to delete the custom model, obtain a new API key, then create a new custom model, just so we can add a new model. very annoying.


    ✅ solution: adding a simple refresh button for custom model to fetch latest model list from OpenRouter, so we can enable them.

  • Tan Duck

    •

    Nov 29, 2025

    I strongly agree and I would love to have that.
    I only use my own OpenRouter API key, I never use free models but I just want to use the latest models and also use TypingMind to try out different models.

    While almost twice a week new models appear in the official list in TypingMind and I have to use the toggle to hide them since I don’t use Anthropic/OpenAI/Google keys, adding new models from OpenRouter is complex. I workaround it by duplicating an existing model and change the parameters.

    By now there are several other OpenRouter like services (NanoGPT, Vercel AI Gateway, Vertex AI Garden, …), maybe they use all the same API and things could be generalised to make live for customers using AI Gateway services easier.

    • Aquamarine Camel

      •

      Mar 24

      yes, it’s very annoying needed to create a new API key, then recreate a custom model every single time a new model is released. we should be able to refresh the models list available from OpenRouter, and toggle on the new models as needed.