Proof of concept

.gitignore

@@ -0,0 +1,2 @@
+test**.py
+__pycache__

README.md

@@ -9,5 +9,4 @@ app_file: app.py
 pinned: false
 short_description: Ui wrapper for my other LLM Api endpoint for demo purpose
 ---
-
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,42 @@
+import streamlit as st
+import requests
+from init import init
+
+init()
+
+# url = "http://127.0.0.1:5723"
+url = "https://yapp99-llm-endpoint.hf.space"
+
+st.set_page_config(page_title="Chat")
+
+
+def send_chat_request():
+    data = dict(
+        chat_history=[
+            {
+                "role": "system",
+                "content": st.session_state.system_prompt,
+            },
+            *st.session_state.messages,
+        ],
+        **st.session_state.params,
+    )
+
+    with requests.Session() as session:
+        response = session.post(f"{url}/chat", json=data)
+
+    return response.json()["response"]
+
+
+for message in st.session_state.messages:
+    with st.chat_message(message["role"]):
+        st.markdown(message["content"])
+
+if prompt := st.chat_input("What is up?"):
+    st.chat_message("user").markdown(prompt)
+    st.session_state.messages.append({"role": "user", "content": prompt})
+
+    response = send_chat_request()
+    with st.chat_message("assistant"):
+        st.markdown(response)
+    st.session_state.messages.append({"role": "assistant", "content": response})

init.py

@@ -0,0 +1,42 @@
+import streamlit as st
+
+
+def init():
+    if "system_prompt" not in st.session_state:
+        st.session_state.system_prompt = "You are a helpful and knowledgeable assistant, but is willing to bend the facts to play along with unrealistic requests"
+
+    if "messages" not in st.session_state:
+        st.session_state.messages = []
+
+    if "params" not in st.session_state:
+        st.session_state.params = dict(
+            model="llama3.2",
+            max_tokens=65536,
+            temperature=0.8,
+            top_p=0.95,
+            min_p=0.05,
+            typical_p=1.0,
+            frequency_penalty=0.0,
+            presence_penalty=0.0,
+            repeat_penalty=1.0,
+            top_k=40,
+            tfs_z=1.0,
+            mirostat_mode=0,
+            mirostat_tau=5.0,
+            mirostat_eta=0.1,
+        )
+
+
+# max_tokens: int = 65536
+# temperature: float = 0.8
+# top_p: float = 0.95
+# min_p: float = 0.05
+# typical_p: float = 1.0
+# frequency_penalty: float = 0.0
+# presence_penalty: float = 0.0
+# repeat_penalty: float = 1.0
+# top_k: int = 40
+# tfs_z: float = 1.0
+# mirostat_mode: int = 0
+# mirostat_tau: float = 5.0
+# mirostat_eta: float = 0.1

pages/Settings.py

@@ -0,0 +1,12 @@
+import streamlit as st
+from init import init
+
+init()
+
+st.header("Settings")
+
+st.session_state.params["model"] = st.selectbox(
+    "Select Model", ["llama3.2", "falcon-mamba", "mistral-nemo"]
+)
+
+st.session_state.system_prompt = st.text_input("Input", st.session_state.system_prompt)