Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
Portex Datalab is a unified platform for frontier evaluations and datasets where domain experts author blinded, real-world tasks and manage pricing, enabling AI labs to test models on truly out-of-sample questions. It hosts benchmarks drawn from real-world occupations in the US SOC system across fields such as chemistry, radiology, data science, and engineering (e.g., SMILES recognition, pneumothorax detection, Polars eval, MRI coil design). Users can create and publish new evals, browse a leaderboard of model scores, and compare performance across industry-relevant tasks.
Parse Score