diff --git a/docs/docs/configuration/object_detectors.md b/docs/docs/configuration/object_detectors.md
index 5c8d70331..2f9c74e60 100644
--- a/docs/docs/configuration/object_detectors.md
+++ b/docs/docs/configuration/object_detectors.md
@@ -357,9 +357,6 @@ detectors: # required
     # 0 means choose automatically
     # increase for better performance if you have a multicore NPU e.g. set to 3 on rk3588
     num_cores: 0
-    # delete old models from config/model_cache/rknn_cache/ folder
-    # set to false to skip, but keep in mind to manually remove unused models
-    purge_model_cache: true
 
 model: # required
   # name of model (will be automatically downloaded) or path to your own .rknn model file
@@ -397,5 +394,5 @@ $ cat /sys/kernel/debug/rknpu/load
 :::
 
 - By default the rknn detector uses the yolonas_s model (`model: path: default-fp16-yolonas_s`). This model comes with the image, so no further steps than those mentioned above are necessary and no download happens.
-- The other choices are automatically downloaded and stored in the folder `config/model_cache/rknn_cache`.
-- Finally, you can also provide your own `.rknn` model. To convert a model to `.rknn` format see the `rknn-toolkit2` (requires a x86 machine). Note, that there is only post-processing for the supported models.
+- The other choices are automatically downloaded and stored in the folder `config/model_cache/rknn_cache`. After upgrading Frigate, you should remove older models to free up space.
+- Finally, you can also provide your own `.rknn` model. You should not save your own models in the `rknn_cache` folder, store them directly in the `model_cache` folder or another subfolder. To convert a model to `.rknn` format see the `rknn-toolkit2` (requires a x86 machine). Note, that there is only post-processing for the supported models.
diff --git a/frigate/detectors/plugins/rknn.py b/frigate/detectors/plugins/rknn.py
index 3eff9afe3..9f6d3a61c 100644
--- a/frigate/detectors/plugins/rknn.py
+++ b/frigate/detectors/plugins/rknn.py
@@ -37,14 +37,6 @@ class Rknn(DetectionApi):
         core_mask = 2**config.num_cores - 1
         soc = self.get_soc()
 
-        if config.purge_model_cache:
-            self.purge_model_cache()
-        else:
-            logger.warning(
-                "Purging model chache is disabled. Remember to manually delete unused models from "
-                + str(model_chache_dir[1:])
-            )
-
         model_props = self.parse_model_input(config.model.path, soc)
 
         from rknnlite.api import RKNNLite
@@ -60,15 +52,6 @@ class Rknn(DetectionApi):
     def __del__(self):
         self.rknn.release()
 
-    def purge_model_cache(self):
-        if os.path.isdir(model_chache_dir):
-            for file in os.listdir(model_chache_dir):
-                if os.path.isfile(file):
-                    if file.endswith("-v2.0.0-1.rknn"):
-                        continue
-                    else:
-                        os.remove(file)
-
     def get_soc(self):
         try:
             with open("/proc/device-tree/compatible") as file: