Show HN: A new benchmark for testing LLMs for deterministic outputs

· Hacker News

Read full story at source