import os import streamlit as st from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline os.environ["TOKENIZERS_PARALLELISM"] = "false" st.title("💡 DeepSeek-Coder-1.3B Playground (CPU Compatible)") st.markdown(""" This playground uses a smaller DeepSeek model that can run on a CPU. You can enter a prompt below and generate completions using Hugging Face Transformers. """) model_id = "deepseek-ai/DeepSeek-Coder-1.3B-base" @st.cache_resource def load_model(): tokenizer = AutoTokenizer.from_pretrained(model_id) model = AutoModelForCausalLM.from_pretrained(model_id) generator = pipeline("text-generation", model=model, tokenizer=tokenizer) return generator generator = load_model() prompt = st.text_area("✍️ Enter your prompt:", "Explain recursion in simple terms.") max_tokens = st.slider("🔧 Max tokens to generate", 16, 256, 64) if st.button("🚀 Generate"): with st.spinner("Generating response..."): result = generator(prompt, max_new_tokens=max_tokens, do_sample=True) st.markdown("### 📜 Output:") st.code(result[0]["generated_text"], language="text")