---
title: Troubleshooting RAG Playground
description: 
url: https://www.elastic.co/elastic/docs-builder/docs/3727/solutions/elasticsearch-solution-project/playground-troubleshooting
products:
  - Kibana
applies_to:
  - Elastic Cloud Serverless: Deprecated
  - Elastic Stack: Deprecated since 9.4, Beta from 9.1 to 9.3, Preview in 9.0
---

# Troubleshooting RAG Playground
<warning>
  **Playground has been deprecated** as of version 9.4. Use [Agent Builder](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/elastic-agent-builder) (generally available since 9.3) instead.With Agent Builder you can:
  - Interact with your Elasticsearch data using a [chat interface](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/chat) in full-screen and sidebar modes
  - Use [built-in agents](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/builtin-agents-reference), [tools](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/tools/builtin-tools-reference), and [skills](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/builtin-skills-reference) out of the box
  - Create [custom agents](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/custom-agents), [custom tools](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/tools/custom-tools), and [custom skills](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/custom-skills)
  - Integrate via [REST API](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/kibana-api) and [MCP](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/tools/mcp-tools)
  - Combine LLM reasoning with [Elastic Workflows](https://www.elastic.co/elastic/docs-builder/docs/3727/explore-analyze/ai-features/agent-builder/agents-and-workflows) to automate complex processes
</warning>

<definitions>
  <definition term="Dense vectors are not searchable">
    Embeddings must be generated using the [inference processor](https://docs-v3-preview.elastic.dev/elastic/docs-builder/docs/3727/reference/ingest-processor/inference-processor) with an ML node.
  </definition>
  <definition term="Context length error">
    You’ll need to adjust the size of the context you’re sending to the model. Refer to [Optimize model context](https://www.elastic.co/elastic/docs-builder/docs/3727/solutions/elasticsearch-solution-project/playground-context).
  </definition>
  <definition term="LLM credentials not working">
    Under **LLM settings**, use the `wrench` button to edit your GenAI connector settings.
  </definition>
  <definition term="Poor answer quality">
    Check the retrieved documents to verify if they are valid. Adjust your Elasticsearch queries to improve the relevance of the documents retrieved. Refer to [View and modify queries](https://www.elastic.co/elastic/docs-builder/docs/3727/solutions/elasticsearch-solution-project/playground-query).
    You can update the initial instructions to be more detailed. This is called *prompt engineering*. Refer to this [OpenAI guide](https://platform.openai.com/docs/guides/prompt-engineering) for more information.
    You might need to click **⟳ Clear chat** to clear chat history and start a new conversation. If you mix topics, the model will find it harder to generate relevant responses.
  </definition>
</definitions>