Is claude-sonnet-4.5-low good at Structured Output?
claude-sonnet-4.5-low ranks #23 of 43 for Structured Output — strong. The top pick for this task is qwen3.7-max-low.
claude-sonnet-4.5-low on each Structured Output sub-task
| Noisy Structured Output Test | 91.2/100 | #16 |
| Missing & Ambiguous Data | 90.0/100 | #27 |
| Transformation | 90.0/100 | #26 |
| Extraction | 85.7/100 | #21 |
| Schema Adherence | 85.7/100 | #30 |
Real examples, graded
WeakSupport ticket to schema 55/100
“The model correctly extracted all data, adhered to the schema types, and properly used null for the missing assignee. However, it completely failed the negative constraint to omit code fences. Because the raw output includes markdown fences, it cannot be parsed directly as JSON, resulting in a hard zero for both schema validity and no wrapper text.”
Frequently asked
Is claude-sonnet-4.5-low good at Structured Output?
claude-sonnet-4.5-low ranks #23 of 43 models we tested for Structured Output, scoring strong.
What is claude-sonnet-4.5-low's strongest Structured Output skill?
Its best sub-task here is Noisy Structured Output Test.
This page is Spring Prompt, running
We just did this for every model. Do it for your prompt.
The rankings above come from running real tasks through real models and scoring every output. Spring Prompt is that same engine — pointed at your prompt, your test cases, and your definition of good.
- Generate test cases from your prompt — no eval set required to start.
- Compare models side by side with quality, cost and latency in one matrix.
- Optimise the winner until the scores say it's ready to ship.
Prompt × model results
12 test cases · 3 evals