Use OpenRouter Models

Open Router (openrouter.ai) is a new service that gives you access to LLM models that you normally won’t have access to (GPT-4 32K, Google PaLM, Anthropic, etc.)
Here is how to use Open Router models on Typing Mind.
Image without caption

1. Create an account

Go to openrouter.ai and create an account. Open Router allows you to use some model for free, or you can buy more credits to use the premium models.
Image without caption

2. Create an API key

After signing up, go to https://openrouter.ai/keys to create an API key.
Enter β€œTyping Mind” to the name field. Leave the β€œCredit limit” empty to set the limit to unlimited.
Image without caption

3. Add Credits

Go to Account > Settings, scroll down to the credit section to Add Credits so you can use the generated API keys.
Image without caption

4. Add a custom model to Typing Mind

4.1. Direct import

Go to https://www.typingmind.com and click the Model Settings, then click β€œAdd Custom Model”.
Image without caption
Image without caption
Choose β€œImport Open Router”, enter the generated API key at Step 2 and click β€œCheck API Key”
Image without caption
A list of chat models will be shown up, you can pick one or multiple chat models at a time from the list.
Image without caption

4.2. Set up as a custom model

  • Navigate the Manage Models menu
  • Scroll down and click β€œAdd Custom Model”
Image without caption
Enter the following information into the Custom Model form:
  • Name: GPT4 via Open Router (you can set any name you want)
  • Endpoint: https://openrouter.ai/api/v1/chat/completions
  • Model ID: openai/gpt-4-32k (change to your desire your model here, make sure the ID is exactly from Open Router model list)
  • Optionally, you can enable the β€œSupport Plugins” toggle if the model you use support OpenAI Function Calling.
Click β€œAdd Custom Headers” and add the following headers:
  • Authorization: Bearer {{your API key created on step 2}}
  • X-Title: typingmind.com
  • HTTP-Referer: https://www.typingmind.com
Here is what it looks like:
Image without caption
After filling all the fields, click β€œTest”.If everything works fine, you will see the message β€œNice, the endpoint is working!”
Click β€œAdd Model” to complete adding the custom model.

5. Use the model

Now you can select the newly added custom model and start chatting:
Image without caption
That’s all!
Happy chatting!