Promptsy

Output Length Calibration

U

@

·

Calibrate output length expectations using examples.

89 copies0 forks
Calibrate {{model}} output lengths for {{task_types}}.

Length examples:
- Concise task: {{concise_example}} → Expected: 50-100 tokens
- Standard task: {{standard_example}} → Expected: 200-400 tokens
- Detailed task: {{detailed_example}} → Expected: 500-1000 tokens

Analyze output lengths and flag over/under-generation patterns.

Details

Category

Analysis

Use Cases

Length calibrationOutput sizingToken optimization

Works Best With

claude-opus-4.5gpt-5.2gemini-2.0-flash
Created Shared

Create your own prompt vault and start sharing