Interactive space in the Foundry portal where a deployed model can be tried out before any code is written. Users tweak temperature, top_p, max tokens and the system prompt, compare models and export sample code; production traffic does not belong here.
Also called model playground.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Foundry playground in context, with comparison tables and the common traps.
Terms in this definition
- WHERE
Limits a SELECT, UPDATE or DELETE to just the rows meeting a condition. Omit it, and the statement hits every row.
- Model temperature
Sampling setting between 0 and 2 controlling how random the output is. Lower values yield focused, repeatable responses and higher ones more creative text; tune this or Top P, but not both.
- Top P
Also known as nucleus sampling, this inference parameter restricts token selection to the most probable share of the distribution; 0.1 keeps the top 10%. It shapes diversity, not length, and should be tuned instead of temperature, not with it.
- MAX
Gives back the biggest value found in a column holding numbers, dates or text, or compares two expressions and returns the bigger. For TRUE/FALSE values, which this DAX aggregation doesn't accept, turn to MAXA.
- System message
Sent ahead of everything else in a chat request, it gives high-priority instructions and context covering role, rules, tone, format and boundaries. Compliance is not guaranteed, only steered; in chat and fine-tuning data it is the system turn.
Related terms
- Completions playground
Older Foundry playground where you typed a prompt and got text back, tied to the legacy Completions API and models like gpt-35-turbo-instruct. Use the chat playground with the Responses or Chat Completions API for anything new.