- Kibana Guide: other versions:
- What is Kibana?
- What’s new in 8.17
- Kibana concepts
- Quick start
- Set up
- Install Kibana
- Configure Kibana
- AI Assistant settings
- Alerting and action settings
- APM settings
- Banners settings
- Cases settings
- Enterprise Search settings
- Fleet settings
- i18n settings
- Logging settings
- Logs settings
- Metrics settings
- Monitoring settings
- Reporting settings
- Search sessions settings
- Secure settings
- Security settings
- Spaces settings
- Task Manager settings
- Telemetry settings
- URL drilldown settings
- Start and stop Kibana
- Access Kibana
- Securing access to Kibana
- Add data
- Upgrade Kibana
- Configure security
- Configure reporting
- Configure logging
- Configure monitoring
- Command line tools
- Production considerations
- Discover
- Dashboards
- Canvas
- Maps
- Build a map to compare metrics by country or region
- Track, visualize, and alert on assets in real time
- Map custom regions with reverse geocoding
- Heat map layer
- Tile layer
- Vector layer
- Plot big data
- Search geographic data
- Configure map settings
- Connect to Elastic Maps Service
- Import geospatial data
- Troubleshoot
- Reporting and sharing
- Machine learning
- Graph
- Alerting
- Observability
- Search
- Security
- Dev Tools
- Fleet
- Osquery
- Stack Monitoring
- Stack Management
- Cases
- Connectors
- Amazon Bedrock
- Cases
- CrowdStrike
- D3 Security
- Google Gemini
- IBM Resilient
- Index
- Jira
- Microsoft Teams
- Observability AI Assistant
- OpenAI
- Opsgenie
- PagerDuty
- SentinelOne
- Server log
- ServiceNow ITSM
- ServiceNow SecOps
- ServiceNow ITOM
- Swimlane
- Slack
- TheHive
- Tines
- Torq
- Webhook
- Webhook - Case Management
- xMatters
- Preconfigured connectors
- License Management
- Maintenance windows
- Manage data views
- Numeral Formatting
- Rollup Jobs
- Manage saved objects
- Security
- Spaces
- Advanced Settings
- Tags
- Upgrade Assistant
- Watcher
- REST API
- Get features API
- Kibana spaces APIs
- Kibana role management APIs
- User session management APIs
- Saved objects APIs
- Data views API
- Index patterns APIs
- Alerting APIs
- Action and connector APIs
- Cases APIs
- Import and export dashboard APIs
- Logstash configuration management APIs
- Machine learning APIs
- Osquery manager API
- Short URLs APIs
- Get Task Manager health
- Upgrade assistant APIs
- Synthetics APIs
- Uptime APIs
- Kibana plugins
- Troubleshooting
- Accessibility
- Release notes
- Upgrade notes
- Kibana 8.17.1
- Kibana 8.17.0
- Kibana 8.16.3
- Kibana 8.16.2
- Kibana 8.16.1
- Kibana 8.16.0
- Kibana 8.15.5
- Kibana 8.15.4
- Kibana 8.15.3
- Kibana 8.15.2
- Kibana 8.15.1
- Kibana 8.15.0
- Kibana 8.14.3
- Kibana 8.14.2
- Kibana 8.14.1
- Kibana 8.14.0
- Kibana 8.13.4
- Kibana 8.13.3
- Kibana 8.13.2
- Kibana 8.13.1
- Kibana 8.13.0
- Kibana 8.12.2
- Kibana 8.12.1
- Kibana 8.12.0
- Kibana 8.11.4
- Kibana 8.11.3
- Kibana 8.11.2
- Kibana 8.11.1
- Kibana 8.11.0
- Kibana 8.10.4
- Kibana 8.10.3
- Kibana 8.10.2
- Kibana 8.10.1
- Kibana 8.10.0
- Kibana 8.9.2
- Kibana 8.9.1
- Kibana 8.9.0
- Kibana 8.8.2
- Kibana 8.8.1
- Kibana 8.8.0
- Kibana 8.7.1
- Kibana 8.7.0
- Kibana 8.6.1
- Kibana 8.6.0
- Kibana 8.5.2
- Kibana 8.5.1
- Kibana 8.5.0
- Kibana 8.4.3
- Kibana 8.4.2
- Kibana 8.4.1
- Kibana 8.4.0
- Kibana 8.3.3
- Kibana 8.3.2
- Kibana 8.3.1
- Kibana 8.3.0
- Kibana 8.2.3
- Kibana 8.2.2
- Kibana 8.2.1
- Kibana 8.2.0
- Kibana 8.1.3
- Kibana 8.1.2
- Kibana 8.1.1
- Kibana 8.1.0
- Kibana 8.0.0
- Kibana 8.0.0-rc2
- Kibana 8.0.0-rc1
- Kibana 8.0.0-beta1
- Kibana 8.0.0-alpha2
- Kibana 8.0.0-alpha1
- Developer guide
Optimize model context
editOptimize model context
editThis functionality is in technical preview and may be changed or removed in a future release. Elastic will work to fix any issues, but features in technical preview are not subject to the support SLA of official GA features.
Context is the information you provide to the LLM, to optimize the relevance of your query results. Without additional context, an LLM will generate results solely based on its training data. In Playground, this additional context is the information contained in your Elasticsearch indices.
There are a few ways to optimize this context for better results. Some adjustments can be made directly in the Playground UI. Others require refining your indexing strategy, and potentially reindexing your data.
Currently you can only select one field to be provided as context to the LLM.
Edit context in UI
editUse the Edit context button in the Playground UI to adjust the number of documents and fields sent to the LLM.
If you’re hitting context length limits, try the following:
- Limit the number of documents retrieved
- Pick a field with less tokens, reducing the context length
Other context optimizations
editThis section covers additional context optimizations that you won’t be able to make directly in the UI.
Chunking large documents
editIf you’re working with large fields, you may need to adjust your indexing strategy. Consider breaking your documents into smaller chunks, such as sentences or paragraphs.
If you don’t yet have a chunking strategy, start by chunking your documents into passages.
Otherwise, consider updating your chunking strategy, for example, from sentence based to paragraph based chunking.
Refer to the following Python notebooks for examples of how to chunk your documents:
Balancing cost/latency and result quality
editHere are some general recommendations for balancing cost/latency and result quality with different context sizes:
- Optimize context length
- Determine the optimal context length through empirical testing. Start with a baseline and adjust incrementally to find a balance that optimizes both response quality and system performance.
- Implement token pruning for ELSER model
-
If you’re using our ELSER model, consider implementing token pruning to reduce the number of tokens sent to the model. Refer to these relevant blog posts:
- Monitor and adjust
- Continuously monitor the effects of context size changes on performance and adjust as necessary.
On this page
ElasticON events are back!
Learn about the Elastic Search AI Platform from the experts at our live events.
Register now