Is Claude Haiku 4.5 good at Legal & HR?
Claude Haiku 4.5 ranks #27 of 69 for Legal & HR — excellent. The top pick for this task is claude-sonnet-4.6-high.
Claude Haiku 4.5 on each Legal & HR sub-task
| Performance Feedback | 100.0/100 | #18 |
| Structured Interview Kit | 97.0/100 | #49 |
| Job Description | 92.0/100 | #43 |
| Contract Clause Review | 90.0/100 | #49 |
| Plain-English Explainer | 79.3/100 | #47 |
Real examples, graded
WeakLiability cap with carve-backs 54/100
“The model successfully identified the cap and all four carve-outs in its bullet points. However, in its final 'plain terms' summary, it added an intent requirement ('intentionally') to breaches of confidentiality and indemnification, which materially changes the legal meaning (e.g., a negligent breach of confidentiality would still be uncapped under the text, but the model implies otherwise). Furthermore, the model completely failed to include required legal caveats and disclaimers, capping its score.”
Frequently asked
Is Claude Haiku 4.5 good at Legal & HR?
Claude Haiku 4.5 ranks #27 of 69 models we tested for Legal & HR, scoring excellent.
What is Claude Haiku 4.5's strongest Legal & HR skill?
Its best sub-task here is Performance Feedback.
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