Selene API

Selene API is an AI evaluation model from Atla AI. It uses LLM-as-a-Judge techniques to score AI applications and return feedback you can act on. Pre-built metrics cover relevance, correctness, and helpfulness, and you can define custom criteria through the Alignment Platform to fit your use case. Typical uses include detecting hallucinations in retrieval-augmented generation apps, comparing answers against ground truth, and checking logical coherence. The site presents the model as accurate and reliable across evaluation benchmarks, and it integrates with existing workflows. A free trial is available, and pricing is usage-based.

- Score AI application output automatically

- Define custom evaluation criteria on the Alignment Platform

- Use pre-built metrics for relevance and correctness

- Detect hallucinations and compare against ground truth

- AI developers building LLM-powered features

- Machine learning and platform teams shipping models

- QA engineers who need repeatable quality checks

Pricing is usage-based, so costs grow with evaluation volume and you should estimate before wiring it into a large test suite. Integration work is needed before it produces value, and benchmark claims from any vendor should be verified on your own data and prompts. Treat the scores as one input into a release decision rather than the only gate. Run one full task end to end first, then expand once you are comfortable with the workflow and the result quality. Behavior may vary across browsers and devices, so test before production use and consult the help docs if errors appear. Name and archive exported files consistently so you can find and reuse them, and keep a backup copy of key data.