Themeisle Docs

Appearance

Troubleshooting Hyve answer quality ​

If Hyve gives incomplete or irrelevant replies, or does not include useful source links, use this guide to diagnose and fix the most common causes.

In this article

Hyve gives incomplete answers ​

Incomplete answers usually happen when your Knowledge Base is not fully processed, the AI connection fails, your content is hard to match semantically, or the selected model is too limited.

Hyve can run in one of two modes, and some steps below depend on which one you use:

  • Hyve Connect (hosted): Hyve handles the AI for you, so you do not need your own OpenAI API key.
  • Self-hosted: Hyve uses your own OpenAI API key for chat and embeddings.

To check which mode is active, open Hyve > Settings > AI > Hyve Connect.

1. Confirm your Knowledge Base finished processing ​

When you add or update sources, Hyve needs time to process them before they can be used in answers.

  1. In your WordPress dashboard, open Hyve > Knowledge Base.
  2. Open Needs attention to see sources that are still pending or waiting to be updated.
  3. Process any outdated items, then test the same question again.

In self-hosted mode, processing runs through WordPress cron tasks, which typically run about once per hour.

📝 Info: If your site has low traffic, WP-Cron may run less frequently, which can delay Knowledge Base updates.

In Hyve Connect mode, your sources are synced to the hosted service instead. If your Knowledge Base is larger than your plan allows, the sync stops and Hyve > Settings > AI > Hyve Connect shows which sources could not be synced. Reduce the number of sources or upgrade your plan, and the sync resumes.

2. Check your AI connection ​

If Hyve cannot reach the AI service, it cannot generate embeddings or reliable responses. Follow the steps for your mode.

In Hyve Connect mode:

  1. Open Hyve > Settings > AI > Hyve Connect.
  2. Check the status. Connected means the hosted service is reachable, and Offline means it is not responding.
  3. Review the message shown next to the status. Hyve reports the reason it cannot answer, such as reaching your plan limit, too many requests in a short time, a Knowledge Base that is still being prepared, or the hosted AI being temporarily unavailable.
  4. If you reached your plan limit, upgrade your plan or wait for the next usage period. For the other states, wait a moment and try again.

In self-hosted mode:

  1. Open Hyve > Settings > AI > Provider & model.
  2. Confirm that API key shows Connected. Not connected means the saved key was rejected, and Unsaved means you have changes you have not saved yet.
  3. Verify in your OpenAI account that API usage is enabled and credit is available.
  4. Return to Hyve > Dashboard and review any visible error notices.

⚠️ Important: If Hyve shows a notice that its encryption keys have changed, your saved credentials can no longer be read and the connection fails until you re-enter them. Re-save your OpenAI API key in Hyve > Settings > AI > Provider & model, re-save your Qdrant connection settings in Hyve > Settings > Integrations > Qdrant if you use Qdrant, and regenerate your tokens in Hyve > Settings > Integrations > API access.

3. Use a more capable assistant model ​

Lower-cost models can return weaker or shorter responses for complex questions.

  1. Open Hyve > Settings > AI > Provider & model.
  2. In Model, select a more capable option. Hyve describes each model in the dropdown, so you can compare them before choosing. The default is a fast, low-cost model that suits most chats, while the higher tiers are aimed at complex questions and cost more per reply.
  3. Save changes and test the same question again to compare answer quality.

📝 Info: This step applies to self-hosted mode only. In Hyve Connect mode the hosted service chooses the model, and the OpenAI settings on this page stay inactive.

💡 Tip: Keep your Temperature and Top P defaults while troubleshooting. Change only one variable at a time so you can clearly see what improves results.

⚠️ Warning: More capable models consume tokens faster and may hit your OpenAI account limits sooner. If the chatbot answers correctly at first but then stops responding after several turns in the same conversation, see Hyve shows "Sorry" or stops responding after several messages below.

4. Write Custom Data for semantic matching ​

Hyve retrieves answers by semantic similarity. The title of an entry is included in what Hyve matches against, so a question that appears only in the title can still be found. A descriptive answer body and natural wording variants still improve retrieval, because your visitor's wording has more to match against.

When you add Custom Data or FAQ-style content, include:

  • the exact question users ask
  • one or more natural-language variants
  • a direct answer in plain language
  • the full URL when you want Hyve to include a documentation link in its reply

Instead of a short generic entry like this:

text
Title: Refunds
Body: We offer refunds.

use a richer entry like this:

text
Title: How can I request a refund?
Body:
How can I request a refund?
Can I get my money back if the plugin did not fit my needs?
To request a refund, follow the steps in this guide:
https://docs.themeisle.com/billing/how-to-request-a-refund

Hyve can only return links that exist in the matched Knowledge Base content. If the source text does not include a URL or clear link target, Hyve may answer without a clickable documentation reference.

To improve link inclusion:

  1. Add the destination URL directly in your source article, Custom Data, or FAQ content.
  2. Re-process any source listed under Needs attention so Hyve stores the latest version.
  3. Ask your test question again and check whether the reply now includes the link.

6. Retest after processing finishes ​

After any Knowledge Base update, run this quick checklist before re-testing:

  1. Open Hyve > Knowledge Base and confirm no relevant items are still pending.
  2. Check Needs attention and process outdated items.
  3. Wait for background processing to finish.
  4. Open Hyve > Dashboard and review any current error notices.
  5. Ask the same question again in the chat and compare the result.

Hyve shows "Sorry" or stops responding after several messages ​

If the chatbot answers correctly at first but then fails after several turns in the same conversation, it is usually being rate limited rather than misconfigured. Hyve sends the conversation history with each request, so token consumption grows as the chat gets longer.

The message your visitors see depends on your mode. In self-hosted mode the chat returns a "Sorry" message. In Hyve Connect mode it returns a short notice that the assistant is temporarily unavailable, and the specific reason is shown to you in the WordPress dashboard rather than to your visitors.

Clearing the chat, refreshing the page, or switching browsers can temporarily restore responses because it resets the conversation history and reduces the token count sent with the next request.

Check the reported reason in Hyve Connect mode ​

  1. Open Hyve > Settings > AI > Hyve Connect.
  2. Review the status and the reason shown next to it, along with your chat usage against your plan.
  3. If you reached your plan limit, upgrade your plan or wait for the next usage period. If the service reports too many requests, wait a moment and try again.

Switch to a lower-cost model in self-hosted mode ​

Lower-cost models consume your account limits more slowly, which makes them better suited to sites with long or high-volume chat sessions.

  1. Open Hyve > Settings > AI > Provider & model.
  2. In Model, select a lower-cost option. The dropdown describes which models are built for high-volume chats.
  3. Save changes and start a new chat session.
  4. Test with a conversation of similar length to the one that triggered the error.

Upgrade your OpenAI usage tier ​

In self-hosted mode, your OpenAI account sets how many tokens per minute your key may use, and more capable models consume that allowance faster. If your site has high chat volume with long conversation threads, raising your usage tier gives your key more headroom. Review your current limits at platform.openai.com.

Was this helpful?