senols

Hi @zargarov,

I just tested it myself by selecting the DeepSeek free version here: https://openrouter.ai/deepseek/deepseek-r1:free, and it worked for me.

Could you let me know which module you’re using (Content Writer, Chatbot, AI Forms, etc.)?

Hi @hebhansen Unfortunately not yet ?? sorry about that ??

Dear Daniel,

Apologies for the late reply, and thank you for reaching out.

You can use AI Forms to create any type of form, including a business name generator. For more information, please check out this guide: https://docs.aipower.org/docs/ai-forms and here some demos: https://aipower.org/ai-form-generator/

Best,

Hello @nytlaegejob

May I know what the issue is?

Hi @raluhh

I believe this issue is related to Google, not the plugin.

I’ve seen some similar posts on Google forums where people are suddenly receiving that message.

Please check this:

https://www.googlecloudcommunity.com/gc/AI-ML/Gemini-API-suddenly-reporting-user-location-is-not-supported-for/m-p/797128

https://discuss.ai.google.dev/t/generate-object-error-user-location-is-not-supported-for-the-api-use/37440

• This reply was modified 2 months, 2 weeks ago by senols.

Hi @tibbie

Does the issue occur with the same product each time, or does it vary?

Could you also clarify at what point it fails—does it stop midway, for example after processing 50 items, or does it fail right at the beginning?

Additionally, please share details about your provider, model, AI parameters, and the buffer time (in seconds).

Hi @webguru4u

You are absolutely correct. I will check this.

Best,

Hi @hebhansen

Seems like U2xkTD5f_2_wpaicg_chattokens table is missing in your WP database. This table is required for the plugin’s chat token management but it is strange that it got deleted.

Are you still getting this error?

Hi @tibbie

Unfortunately, our plugin doesn’t currently have the feature to run it in the background.

Best,

Hi @tibbie

The speed of generating descriptions depends on factors like network, API endpoint, and the AI model used. However, 15–20 seconds is longer than normal—it usually takes just a few seconds.

Could you share which AI provider and model you’re using?

Hi @keytrackde

At the moment, our plugin doesn’t have any integration with phone systems.

However, we do have plans to explore integrations with platforms like WhatsApp and Zapier ??

Have a good day

Hi @hebhansen

To make chatbots consider site categories, you can use a vector database and embeddings.

When content from your site is embedded into the vector database via our plugin, it includes fields such as post title, post content, post URL, and post category. However, tags are not embedded currently, but I can consider including tags in future releases.

You can also instruct your bot to check for the category information provided in the context and use it to give more relevant responses.

Hi @keytrackde

Currently, all three embedding models (text-embedding-ada-002, text-embedding-3-small, and text-embedding-3-large) have a maximum input size of 8191 tokens.

This means data exceeding this token limit cannot be embedded directly.

If you only have a few large posts, here are a couple of workarounds:

Split the Content: Break large posts into smaller sections and embed them manually using the Knowledge Builder text field. You can copy and paste the content in two or three parts.
Use a PDF Converter: Convert your large posts into PDFs using an online tool, then upload them with our plugin’s PDF upload feature. The plugin will embed the content page by page.

From our testing, we’ve found that text-embedding-3-small is faster and cheaper but less accurate compared to text-embedding-ada-002 and text-embedding-3-large. For better accuracy, we’ve migrated all our posts to text-embedding-ada-002.

Let me know if you have any questions or need further help.`

Hi @hebhansen

Apologies for the delayed response.

The issue is unlikely to be related to your keyboard, as cache and keyboard inputs don’t interact.

However, caching plugins can sometimes cache nonces that we use in the chatbot for security reasones, causing this error.

Another user fixed a similar issue by adding exceptions for nonces in their caching plugin.

You can check their solution here: https://www.ads-software.com/support/topic/error-nonce-verification-failed/#post-17465491.

I am not sure but this might help fix your issue as well.

Hi @hebhansen,

Thank you for your kind words. I’m glad you like the new backend—I put a lot of effort into implementing it, so it’s great to hear your feedback. ??

I’m aware of the current incompatibility with multisite setups and already know how to fix it and I plan to fix it soon.

Re: log access: I’ll take this into consideration for future updates.

Regarding your concerns about training data loss: training data is stored independently, so it won’t be affected by theme or model changes. The only scenarios where training data might be lost are if the database becomes corrupted, deleted, or if the vector database is manually removed. Otherwise, your data should remain safe and functional.

At the moment, though, I’m fully focused on integrating the OpenAI Assistant API. I believe this integration is essential as it will allow you to create your own assistant on OpenAI’s platform and use it directly within the plugin.

Once the assistant integration is complete, I’ll shift my focus to fix multisite issues and improving AI forms (including implementing markdown parsing libraries).

As I’m the only developer working on this plugin, I truly appreciate your patience and understanding. Thank you for your support.