-
feat: wire kompress_model through compress pipeline
发布于
2026-04-17 18:59:29 +00:00 CompressConfig.kompress_model was declared but not passed to transforms.
Now it flows through pipeline.apply() to ContentRouter, enabling:- Custom Kompress models via HuggingFace model ID
- kompress_model="disabled" to skip ML compression entirely
(SmartCrusher + CacheAligner still run) - Default behavior (kompress_model=None) unchanged
下载附件