AI Playground: Prototyping and Testing LLMs
Who this is for:
Architecture / Concept Overview: AI Playground: Prototyping and Testing LLMs
AI Playground connects directly to Foundation Model APIs and Model Serving endpoints, giving users instant access to multiple LLMs.
%%{init: {"theme":"base","themeVariables":{"background":"#0B0E14","primaryTextColor":"#E0E6ED","lineColor":"#5D6470","darkMode":true,"primaryColor":"#2E4A4A","secondaryColor":"#374151","secondaryTextColor":"#E0E6ED","tertiaryColor":"#111827","tertiaryTextColor":"#E0E6ED","edgeLabelBackground":"#1f2937"}}}%%
flowchart LR
classDef source fill:#3F4B59,stroke:#9CA3AF,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef ingestion fill:#5A4B36,stroke:#C9A86B,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef processing fill:#535072,stroke:#8E82B4,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef storage fill:#2E4A4A,stroke:#5FAFA8,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef serving fill:#3D5550,stroke:#6BB7AA,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef governance fill:#5A3F52,stroke:#C28BB0,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
USER[User] -->|Interact| PLAY[AI Playground UI]
PLAY -->|Select| FM_A[Model A - Llama]
PLAY -->|Select| FM_B[Model B - DBRX]
FM_A -->|Query| API_A[Foundation Model API]
FM_B -->|Query| API_B[Foundation Model API]
PLAY -->|Export| NB[Notebook Code]
PLAY -->|Compare| SIDE[Side-by-Side Results]
USER:::source
PLAY:::ingestion
FM_A:::processing
FM_B:::processing
API_A:::serving
API_B:::serving
NB:::governance
SIDE:::storage
*AI Playground flow: users select models, compare outputs side-by-side, and export configurations as code.*
%%{init: {"theme":"base","themeVariables":{"background":"#0B0E14","primaryTextColor":"#E0E6ED","lineColor":"#5D6470","darkMode":true,"primaryColor":"#2E4A4A","secondaryColor":"#374151","secondaryTextColor":"#E0E6ED","tertiaryColor":"#111827","tertiaryTextColor":"#E0E6ED","edgeLabelBackground":"#1f2937"}}}%%
graph TD
classDef source fill:#3F4B59,stroke:#9CA3AF,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef ingestion fill:#5A4B36,stroke:#C9A86B,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef processing fill:#535072,stroke:#8E82B4,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef storage fill:#2E4A4A,stroke:#5FAFA8,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef serving fill:#3D5550,stroke:#6BB7AA,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
classDef governance fill:#5A3F52,stroke:#C28BB0,stroke-width:2px,rx:8,ry:8,color:#E0E6ED
PLAY[AI Playground] --> CHAT[Chat Mode]
PLAY --> COMPLETE[Completion Mode]
PLAY --> COMPARE[Compare Mode]
CHAT --> SYS[System Prompt]
CHAT --> PARAMS[Parameters - temp, max_tokens]
CHAT --> HISTORY[Conversation History]
COMPARE --> LEFT[Model A Output]
COMPARE --> RIGHT[Model B Output]
PLAY:::governance
CHAT:::processing
COMPLETE:::storage
COMPARE:::serving
SYS:::ingestion
PARAMS:::ingestion
HISTORY:::source
LEFT:::source
RIGHT:::source
*AI Playground modes and configuration options.*
Key Terms
Prerequisites and Setup
- Databricks workspace with Foundation Model APIs enabled.
- User access to the Machine Learning persona in the workspace UI.
- No cluster or code required — AI Playground is serverless.
Step-by-Step Implementation
Configuration Reference
| Parameter | Default | Description |
|---|---|---|
temperature | 1.0 | Controls randomness in token sampling |
max_tokens | 256 | Maximum number of tokens in the response |
top_p | 1.0 | Nucleus sampling probability threshold |
system_prompt | "" | System message defining model behaviour |
stop_sequences | [] | Tokens that cause the model to stop generating |