A collection of notebooks demonstrating Databricks features and integrations.
| Notebook | Description |
|---|---|
| ai_sql_functions | Query Foundation Models with SQL using ai_query |
| fast_classification_provisioned | High-throughput text classification with Provisioned Throughput |
| fm_api_mlflow_prompt_eng | Compare models and prompts with the MLflow Prompt Engineering UI |
| fm_api_openai_sdk | Use the Foundation Model API with the OpenAI Python SDK |
| fm_api_outside_db | Call the Foundation Model API from outside Databricks |
| intro_generation_parameters | Control model behavior with temperature, top_p, top_k, and more |
| manage_chat_sessions | Manage chat sessions with ChatSession and custom history |
| playground | Test and compare LLMs in the Databricks AI Playground |
| sam2-on-databricks | Run SAM2 video object segmentation on Databricks |
| serve_marketplace | Deploy Marketplace models with Delta Sharing and Model Serving |
| streaming_outputs | Stream responses from Foundation Model API models |
| vector_search_fm_api | Set up Vector Search with Foundation Model API embeddings |
| AI Gateway helper notebook | Query system tables for AI serving endpoint usage |
| Connect to more data with Lakehouse Federation | Query external databases via Lakehouse Federation |
| Current LDN 2025 transformWithState | Stateful Structured Streaming with transformWithStateInPandas |
| Data Engineering SQL Holiday Specials | New SQL features from late 2024: materialized views, federation, and more |
| Ham Sandwiches Custom Config | Pass variables between languages and add custom Delta config |
| Notebook | Description |
|---|---|
| 2024-11-27-dspy | Build and optimize LLM programs with DSPy and MLflow |
| gemini-trace-tool | MLflow tracing with Gemini 2.0 Flash and tool calling |
In your Databricks workspace, go to the workspace browser, select Import (from the kebab menu or right-click menu), and paste the raw GitHub URL of the notebook you want to import.
Navigate to Repos in your Databricks workspace, click Add Repo, and enter:
https://github.com/databricks-solutions/devrel-examples
All notebooks are in the notebooks/ directory.