diff --git a/README.md b/README.md
index 40ac2f2..4d2bff8 100644
--- a/README.md
+++ b/README.md
@@ -2,7 +2,14 @@
 
 Code for paper [Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes](https://arxiv.org/abs/2305.02301)
 
-## About this fork
+## Changes in this fork
+* [x] Add support for GCS
+* [x] Add command line invocation with arguments
+* [ ] Add support for hosting distilled models using docker
+* [ ] Add support for hosting models as vertex AI endpoints
+* [ ] Add support for hosting models as TF Serving endpoints
+* [ ] Add [kedro](https://kedro.readthedocs.io/en/stable/) pipeline for distillation
+* [ ] Add support for [vertex AI](https://cloud.google.com/vertex-ai/docs) pipelines
 
 **Work in progress.**
 
@@ -37,7 +44,6 @@ distillm
 ```
 
 #### Example usages
-
 - Distilling step-by-step with `PaLM label` and `PaLM rationale`:
 ```python
 distillm  --from_pretrained google/t5-v1_1-small \
@@ -71,6 +77,7 @@ distillm  --from_pretrained google/t5-v1_1-small \
 - `--output_dir`: The directory for saving the distilled model
 - `--gcs_project`: The GCP project name
 - `--gcs_path`: The GCS path. **_train.json** and **_test.json** will be added to the path
+
 ## Cite
 If you find this repository useful, please consider citing:
 ```bibtex
diff --git a/hosting/app.py b/hosting/app.py
index 9989eac..185dff3 100644
--- a/hosting/app.py
+++ b/hosting/app.py
@@ -11,10 +11,15 @@
 async def home():
   return {"message": "Machine Learning service"}
 
-@router.post("/serve")
+@router.post("/predict")
 async def data(data: dict):
     return {"message": "Data received"}
 
+@router.get("/health")
+async def health():
+  return {"message": "Model is healthy"}
+
+
 app.include_router(router)
 
 if __name__ == "__main__":