README.md

metadata

title: SimpleChatbot
emoji: 🤖
colorFrom: purple
colorTo: pink
sdk: streamlit
sdk_version: 1.31.1
app_file: app.py
pinned: false
license: mit
models:
  - openai/gpt-oss-120b
  - mistralai/Mistral-7B-Instruct-v0.2
  - google/gemma-3-27b-it
  - google/gemma-2-27b-it
  - google/gemma-2-9b-it
  - google/gemma-2-2b-it
  - HuggingFaceH4/zephyr-7b-gemma-v0.1
  - HuggingFaceH4/zephyr-7b-beta
  - meta-llama/Meta-Llama-3-8B-Instruct
  - meta-llama/Meta-Llama-3.1-8B-Instruct
  - deepseek-ai/DeepSeek-R1-Distill-Llama-70B
  - Qwen/Qwen2.5-Coder-32B-Instruct
  - moonshotai/Kimi-K2-Instruct

🤖 SimpleChatbot

SimpleChatbot is a lightweight Streamlit app that lets you chat with multiple large language models (LLMs) from a single interface. It’s built as a simple, interactive way to explore and compare different models without switching tools or setups.

---

✨ Features

• Chat with multiple LLMs from one UI
• Switch models on the fly (conversation resets automatically)
• Adjustable temperature for response creativity
• Built-in rate limiting & cooldowns for fair usage
• Optional Hugging Face access token support
• Session-based logging and usage tracking
• Clean, minimal Streamlit interface

---

🧠 Supported Models

This app supports a mix of general-purpose, reasoning, and coding-focused models, including:

• OpenAI GPT-OSS-120B
• Meta Llama 3 / 3.1
• Google Gemma (2 & 3 series)
• DeepSeek R1 (distilled)
• Qwen 2.5 Coder
• Kimi-K2 Instruct
• Zephyr & Mistral variants

Available models may change depending on backend availability.

---

🎯 Why This Exists

This project was built to:

• Quickly compare LLM behavior across different model families
• Provide a simple reference implementation for multi-model chat apps
• Demonstrate how to build a Streamlit chatbot using modern inference APIs
• Serve as a lightweight demo for experimentation, education, and prototyping

It intentionally avoids complex agent logic or tooling to keep the focus on model responses and UX.

---

🚀 How to Use

1. Select a model client in the sidebar
2. Choose a model
3. Adjust the temperature if desired
4. Start chatting!

Responses are intentionally kept brief to keep interactions fast and readable.

---

⚠️ Notes

• Generated content may be inaccurate or incorrect
• API calls are limited per session to prevent abuse
• If a model fails, it may be temporarily unavailable or updating

---

👤 Author

Created by Nigel Gebodh
🌐 https://ngebodh.github.io/

Learn how to build this chatbot yourself:
👉 https://ngebodh.github.io/projects/2024-03-05/