New AI Text Plugin

[fatihak]

Hey Everyone,

I just pushed the addition of the AI Text Plugin, which allows you to display any text content using Open AI's GPT-4o text models.
To get the latest changes, you should be able to just do a git pull and restart the InkyPi service. Please note that this plugin requires the OPEN_AI_SECRET API key to be configured in a .env file, similar to the setup for the AI Image Plugin.
Let me know if you have any questions, feedback, or run into any issues with the plugin.

Thanks

[Its4Nik]

Hey Everyone,

I just pushed the addition of the AI Text Plugin, which allows you to display any text content using Open AI's GPT-4o text models.
To get the latest changes, you should be able to just do a git pull and restart the InkyPi service. Please note that this plugin requires the OPEN_AI_SECRET API key to be configured in a .env file, similar to the setup for the AI Image Plugin.
Let me know if you have any questions, feedback, or run into any issues with the plugin.

Thanks

just as a question, is Ollama planned? So that we can add self hosted AIs?

[fatihak]

It's currently not planned unless there is more interest. However, based on this article it looks like you can point the open ai python client to your Ollama host so you should be able to get it working by changing a few lines in src/plugins/ai_text/ai_text.py:

--- a/src/plugins/ai_text/ai_text.py
+++ b/src/plugins/ai_text/ai_text.py
@@ -58,7 +58,10 @@ class AIText(BasePlugin):
             raise RuntimeError("Text Prompt is required.")

         try:
-            ai_client = OpenAI(api_key = api_key)
+            ai_client = OpenAI(
+                base_url = 'your_ollama_host_endpoint',
+                api_key = api_key
+            )
             prompt_response = AIText.fetch_text_prompt(ai_client, text_model, text_prompt)
         except Exception as e:
             logger.error(f"Failed to make Open AI request: {str(e)}")
@@ -107,7 +110,7 @@ class AIText(BasePlugin):

         # Make the API call
         response = ai_client.chat.completions.create(
-            model=model,
+            model="llama2", # ollama model
             messages=[```

[verynifty12]

I have my key, does it get placed into a file with a .env extension? Any documentation on exactly what it is looking for? or point me in the right direction? :)

Great program. I have had fun with it

[fatihak]

You can find instructions here or alternatively watch my video on it :)