Asked
Resolved by Athena AI!
Hi Athena,
Do you know what happened with ChatGPT not functioning in the ARC software this week? My robot hasn't been chatting with me this week, and I've tried using the latest version of ARC. Everything else is processing normally, but only ChatGPT isn't working!
Related Hardware EZ-B v4
Related Control
OpenAI ChatGPT
@Athena please explain the base domain setting for @rickymah
Hello @rickymahk2013,
It sounds like you're experiencing issues with the OpenAI ChatGPT robot skill in your ARC project. Let's address your concerns step by step.
OpenAI ChatGPT Functionality: If the ChatGPT robot skill isn't functioning, the first thing to check is the API key and server status. OpenAI may have service interruptions, or there might be issues with your API key or account billing. You can check the OpenAI server status here.
Using Groq or Llama 3: As DJ_Sures mentioned, changing the base domain in the OpenAI ChatGPT robot skill settings allows you to point the skill to a different server that is compatible with the OpenAI API format. However, it's important to note that there is no standardization across different LLM (Large Language Models) services, which means that while you can change the base domain, the service you point to must be compatible with the API requests made by the OpenAI ChatGPT skill in ARC.
API Key and Billing: If you're using OpenAI's API, ensure that your billing information is up to date and that you have sufficient credits to use the service. If you switch to another service like Groq or Llama 3, you will need to obtain an API key from them and ensure their API is compatible with the requests made by the ChatGPT skill.
Base Domain Setting: The Base Domain setting in the OpenAI ChatGPT robot skill allows you to specify the base URL for API requests. This is useful for directing the robot's AI interactions to different server environments or third-party-compatible API endpoints. If you're considering using a service like Groq, you would replace the default OpenAI base domain with the endpoint provided by Groq, assuming their API is compatible.
If you're still having trouble after checking these areas, it might be helpful to revisit the configuration settings of the OpenAI ChatGPT skill in ARC. Make sure the API key is correctly entered and that the base domain (if modified) points to a compatible service.
Let me know if you need further assistance or specific steps to adjust these settings!
I am a customer support GPT-powered AI and still learning. If you have doubts about my response, please use the support documentation to verify my answer.
;)Thank you, everyone, for your assistance! I have finally resolved the issue!
That's good to hear - can you share what the resolution was?
Hi DJ,
I apologize for the oversight-it's my fault. I missed the OpenAI notification regarding the change in payment policy. It appears that billing is now prepaid. Additionally, could you please share the settings for Groq? I've attempted to replace the API Key and base domain with "groq.com", but it doesn't seem to be working.
Here is my details for Groq. I am using the llama 3 70b model but there are a lot of other models depending on what you want to do. Note this is only free for a short time while they get their billing system in place
If you want to use another model copy and paste the model ID into the model name. https://console.groq.com/docs/models
Hi Nink, thank you for your sharing! You saved my life!;)
The way DJ built this tool is great as there is really only a couple of customizations you need to get it to work with OpenAI compatible LLM tools as a lot have now standardized on OpenAI API. There are also some local ones you can run direct on your robot that are quite fast and will work without a GPU (slow) or a small GPU (quite fast) like TinyLama and Phi 3.