Update app.py

app.py

@@ -6,7 +6,7 @@ from threading import Timer
 
 HUGGINGFACE_TOKEN = os.environ.get("HUGGINGFACE_TOKEN")
 def get_available_free():
-    models = InferenceClient().list_deployed_models("text-generation-inference")['text-generation']
+    models = InferenceClient(token=HUGGINGFACE_TOKEN).list_deployed_models("text-generation-inference")['text-generation']
     models_conclusion = {
         "Model": [],
         "API": [],
@@ -87,7 +87,21 @@ def color_status(api_value, cell_value):
 def search_models(query):
     return display_table(query)
 
-description = "This is a space that retrieves the status of all supported HF LLM Serverless Inference APIs.\nUpdates every 2 hours!"
+description = """
+This is a space that retrieves the status of all supported HF LLM Serverless Inference APIs.
+*Updates every 2 hours!*
+
+If you are a student or you just want to quickly see what models are available to experiment for free, you are most likely highly interested on the free API huggingface provides... but like me, you struggle to find what models are available or not!
+This is why I made this space that every 2 hours checks and updates the status of the list of LLMs that are in theory supported by retrieving the list in `InferenceClient().list_deployed_models("text-generation-inference")['text-generation']`.
+
+So all you need is plug:
+```py
+from huggingface_hub import InferenceClient
+inf = InferenceClient(model = "MODEL", token = "TOKEN")
+response = inf.text_generation("And play !!")
+print(response)
+```
+"""
 with gr.Blocks() as demo:
     gr.Markdown("## HF Serverless LLM Inference API Status")
     gr.Markdown(description)