paradite/eval-data

Prompts and evaluation data for LLMs on real world coding and writing tasks

29
/ 100
Experimental

This provides a collection of prompts and expected outputs for evaluating how well large language models (LLMs) perform on various real-world coding and writing tasks. It takes in a specific task scenario (like writing a Next.js todo app or explaining Kanji) and provides benchmark data to assess an LLM's generated code or text. This is designed for AI researchers, prompt engineers, and product managers who are developing or integrating LLM-powered applications.

No commits in the last 6 months.

Use this if you need pre-defined, diverse datasets to systematically test and compare the performance of different LLMs or prompt strategies on practical development and content creation challenges.

Not ideal if you are looking for a tool to generate new code or content directly, rather than evaluate an LLM's output.

LLM-evaluation prompt-engineering AI-benchmarking code-generation-testing content-creation-assessment
No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 13 / 25

How are scores calculated?

Stars

17

Forks

3

Language

TypeScript

License

Last pushed

Sep 13, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/paradite/eval-data"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.