Set base_model to fastino/gliner2.5-multi-v1

Include pinned encoder config

Document ANE placement across all FP16 extraction stages

Add validated W8 extraction packages and ANE profiling for multi

Document materialized Hub downloads for Core ML packages

Place W8 package at runtime model root

Add validated asymmetric embedding W8 classification option

Add validated GLiNER2 extraction Core ML stages

Add Apache-2.0 license text

Publish GLiNER2.5 multilingual Core ML classification exports

initial commit