Point the post-training quantization docs to LiteRT Quantizer.

tensorflow_model_optimization/g3doc/guide/get_started.md

@@ -16,14 +16,13 @@ provide the efficiency required by your application.
 
 ## 3. Post-training tooling
 
-If you cannot use a pre-trained model for your application, try using
-[TensorFlow Lite post-training quantization tools](./quantization/post_training)
-during [TensorFlow Lite conversion](https://www.tensorflow.org/lite/convert),
-which can optimize your already-trained TensorFlow model.
-
-See the
-[post-training quantization tutorial](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/g3doc/performance/post_training_quant.ipynb)
-to learn more.
+If you cannot use a pre-trained model for your application, convert your
+already-trained model to a float32 LiteRT model and quantize it with
+[LiteRT Quantizer](https://github.com/google-ai-edge/LiteRT/tree/main/litert_quantizer).
+
+Note: Post-training quantization in the TensorFlow Lite converter is
+deprecated. See the
+[migration guide](https://developers.google.com/edge/litert/quantization/tflq_to_litertq_migration).
 
 ## Next steps: Training-time tooling
 

tensorflow_model_optimization/g3doc/guide/index.md

@@ -37,7 +37,7 @@ The area of model optimization can involve various techniques:
     and distillation
 
 Our toolkit supports
-[post-training quantization](./quantization/post_training.md),
+[post-training quantization](./quantization/post_training.md) (deprecated),
 [quantization aware training](./quantization/training.md),
 [pruning](./pruning/index.md), and [clustering](./clustering/index.md). The
 toolkit also provides experimental support for
@@ -46,6 +46,15 @@ various techniques.
 
 ### Quantization
 
+Warning: Post-training quantization in the TensorFlow Lite converter is
+deprecated. For post-training quantization, convert your model to a float32
+LiteRT model and quantize it with
+[LiteRT Quantizer](https://github.com/google-ai-edge/LiteRT/tree/main/litert_quantizer).
+See the
+[migration guide](https://developers.google.com/edge/litert/quantization/tflq_to_litertq_migration).
+This doesn't affect the training-time techniques in this toolkit, such as
+pruning, clustering and quantization aware training.
+
 Quantized models are those where we represent the models with lower precision,
 such as 8-bit integers as opposed to 32-bit float. Lower precision is a
 requirement to leverage certain hardware.

tensorflow_model_optimization/g3doc/guide/quantization/post_training.md

@@ -1,5 +1,12 @@
 # Post-training quantization
 
+Warning: Post-training quantization in the TensorFlow Lite converter is
+deprecated. To quantize a model, convert it to a float32 LiteRT model and
+quantize it with
+[LiteRT Quantizer](https://github.com/google-ai-edge/LiteRT/tree/main/litert_quantizer).
+See the
+[migration guide](https://developers.google.com/edge/litert/quantization/tflq_to_litertq_migration).
+
 Post-training quantization includes general techniques to reduce CPU and
 hardware accelerator latency, processing, power, and model size with little
 degradation in model accuracy. These techniques can be performed on an