Comit

Custy-gpt.py

@@ -0,0 +1,31 @@
+import streamlit as st
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+
+# Load the Falcon-7B model with 8-bit quantization
+model_id = "tiiuae/falcon-7b-instruct"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id,
+    load_in_8bit=True,
+    device_map="auto",
+    trust_remote_code=True
+)
+
+# Create a text generation pipeline
+pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
+
+# Streamlit app
+st.title("Falcon-7B Text Generation")
+st.write("Generate text using Falcon-7B with 8-bit quantization.")
+
+prompt = st.text_input("Enter your prompt:")
+if prompt:
+    outputs = pipe(
+        prompt,
+        max_new_tokens=50,
+        temperature=0.7,
+        top_k=50,
+        top_p=0.9,
+        do_sample=True
+    )
+    st.write(outputs[0]["generated_text"])

requirements.txt

@@ -0,0 +1,5 @@
+torch
+transformers
+accelerate
+bitsandbytes
+gradio