lukalabs/cakechat

Health v2 · maintenance only

Flags: archived

Adoption not part of the score

1707 stars · 911 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

CakeChat is a Python backend for building emotional generative chatbots using a hierarchical recurrent encoder-decoder (seq2seq) architecture built on Keras and TensorFlow. It supports conditioning responses on categorical variables such as emotion or persona, and ships with an HTTP server and Telegram bot integration.

Use cases

• build a chatbot that expresses emotions in its replies
• train a seq2seq dialogue model on my own conversation data
• fine-tune a pre-trained neural conversational model
• run a Telegram bot powered by a neural dialogue system
• condition chatbot responses on a persona or other categorical variable
• serve a trained dialogue model over an HTTP API

When to choose

• you specifically need an RNN/HRED-based emotional dialogue system in Python
• you want a self-contained chatbot backend with HTTP and Telegram integrations
• you are studying or reproducing older seq2seq conversational model research

When to avoid

• you need a modern, maintained chatbot - the project is explicitly unmaintained
• you want state-of-the-art dialogue quality - transformer models like DialoGPT are recommended instead
• your stack requires recent TensorFlow/Keras versions - it targets TensorFlow 1.12 and Python 3.5

Facets

library · maturity abandoned

nlp deep-learning chatbot machine-learning llm-training artificial-intelligence chatbots deep-learning python seq2seq dialogue-system emotional-chatbot keras tensorflow hred unmaintained natural-language-processing linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="lukalabs/cakechat")