The PCMDI Metrics Package (PMP) is an open-source Python-based framework that enables objective "quick-look" comparisons and benchmarking of Earth System Models (ESMs) against the latest generation observations. The PMP has been used for routine and systematic evaluation of thousands of simulations from Coupled Model Intercomparison Projects (CMIPs) focused primarily on atmospheric quantities. Ongoing work aims for seamless application of the tool to the upcoming CMIP next generation (i.e., CMIP7), with an aspiration to aid modeling groups during their development cycle. The latest version of PMP offers a diverse suite of evaluation capabilities covering large- to global-scale climatology and annual cycle, variability modes such as tropical and extratropical variability modes e.g., ENSO and MJO, regional monsoons, cloud feedback, and high-frequency characteristics e.g., extremes. Current work is expanding the scope of PMP to include the evaluation of the following: (1) Quasi-Biennial Oscillation (QBO) and its teleconnection to MJO, (2) atmospheric blocking and rivers leveraging Machine Learning based pattern detection algorithms, and (3) polar and high-latitude regions by implementing the sectional sea-ice area metrics. The PMP is also advancing its evaluation capabilities to help evaluate higher-resolution simulations such as those from the HighResMIPs, cloud-resolving E3SM experiments, and regionally downscaled products. This presentation will describe the philosophy of routine model evaluation, introduce the PMP, share progress on current polar metrics, and discuss plans and opportunities to connect with ongoing efforts.