updated build names for documentation

docs/docs/configuration/object_detectors.md

@@ -402,13 +402,13 @@ detectors:
 
 The `rocm` detector allows one to use [ultralytics](https://github.com/ultralytics/ultralytics) yolov8 models on AMD GPUs and iGPUs. You need the special frigate build that contains the AMD/ROCm software stack:
 
-- `ghcr.io/harakas/frigate:latest-rocm`
+- `ghcr.io/blakeblackshear/frigate:stable-rocm`
 
 As the ROCm software stack is quite bloated, there are also smaller versions for specific GPU chipsets:
 
-- `ghcr.io/harakas/frigate:latest-rocm-gfx900`
-- `ghcr.io/harakas/frigate:latest-rocm-gfx1030`
-- `ghcr.io/harakas/frigate:latest-rocm-gfx1100`
+- `ghcr.io/blakeblackshear/frigate:stable-rocm-gfx900`
+- `ghcr.io/blakeblackshear/frigate:stable-rocm-gfx1030`
+- `ghcr.io/blakeblackshear/frigate:stable-rocm-gfx1100`
 
 ### Docker settings for GPU access
 

docs/docs/frigate/installation.md

@@ -150,6 +150,10 @@ The community supported docker image tags for the current stable version are:
 - `stable-tensorrt-jp5` - Frigate build optimized for nvidia Jetson devices running Jetpack 5
 - `stable-tensorrt-jp4` - Frigate build optimized for nvidia Jetson devices running Jetpack 4.6
 - `stable-rk` - Frigate build for SBCs with Rockchip SoC
+- `stable-rocm` - Frigate build for [AMD GPUs and iGPUs](../configuration/object_detectors.md#amdrocm-gpu-detector), all drivers
+  - `stable-rocm-gfx900` - AMD gfx900 driver only
+  - `stable-rocm-gfx1030` - AMD gfx1030 driver only
+  - `stable-rocm-gfx1100` - AMD gfx1100 driver only
 
 ## Home Assistant Addon