In the evolving landscape of local-first AI assistants, the release of Mr. Chatterbox by Trip Venturella marks a significant milestone for the OpenClaw ecosystem. This language model, trained exclusively on out-of-copyright Victorian-era texts from the British Library, embodies the principles of open-source development and agent-centric automation that define platforms like OpenClaw. With a corpus of 28,035 books and 2.93 billion tokens, Mr. Chatterbox offers a glimpse into how niche, historically focused models can integrate into local AI workflows, despite its current limitations in conversational utility.
Mr. Chatterbox is a language model trained entirely from scratch on a corpus of over 28,000 Victorian-era British texts published between 1837 and 1899, drawn from a dataset made available by the British Library. The model has absolutely no training inputs from after 1899 — the vocabulary and ideas are formed exclusively from nineteenth-century literature. Mr. Chatterbox’s training corpus was 28,035 books, with an estimated 2.93 billion input tokens after filtering. The model has roughly 340 million parameters, roughly the same size as GPT-2-Medium. The difference is, of course, that unlike GPT-2, Mr. Chatterbox is trained entirely on historical data.
Given how hard it is to train a useful LLM without using vast amounts of scraped, unlicensed data, I’ve been dreaming of a model like this for a couple of years now. What would a model trained on out-of-copyright text be like to chat with? Thanks to Trip we can now find out for ourselves! The model itself is tiny, at least by Large Language Model standards—just 2.05GB on disk. You can try it out using Trip’s HuggingFace Spaces demo: Honestly, it’s pretty terrible. Talking with it feels more like chatting with a Markov chain than an LLM—the responses may have a delightfully Victorian flavor to them but it’s hard to get a response that usefully answers a question.
The 2022 Chinchilla paper suggests a ratio of 20x the parameter count to training tokens. For a 340m model that would suggest around 7 billion tokens, more than twice the British Library corpus used here. The smallest Qwen 3.5 model is 600m parameters and that model family starts to get interesting at 2b—so my hunch is we would need 4x or more the training data to get something that starts to feel like a useful conversational partner. But what a fun project! Running it locally with LLM I decided to see if I could run the model on my own machine using my LLM framework. I got Claude Code to do most of the work— here’s the transcript. Trip trained the model using Andrej Karpathy’s nanochat, so I cloned that project, pulled the model weights and told Claude to build a Python script to run the model.
Once we had that working (which ended up needing some extra details from the Space demo source code) I had Claude read the LLM plugin tutorial and build the rest of the plugin. llm-mrchatterbox is the result. Install the plugin like this: llm install llm-mrchatterbox. The first time you run a prompt it will fetch the 2.05GB model file from Hugging Face. Try that like this: llm -m mrchatterbox “Good day, sir”. Or start an ongoing chat session like this: llm chat -m mrchatterbox. If you don’t have LLM installed you can still get a chat session started from scratch using uvx like this: uvx –with llm-mrchatterbox llm chat -m mrchatterbox. When you are finished with the model you can delete the cached file using: llm mrchatterbox delete-model.
This is the first time I’ve had Claude Code build a full LLM model plugin from scratch and it worked really well. I expect I’ll be using this method again in the future. I continue to hope we can get a useful model from entirely public domain data. The fact that Trip was able to get this far using nanochat and 2.93 billion training tokens is a promising start. Update 31st March 2026: I had missed this when I first published this piece but Trip has his own detailed writeup of the project which goes into much more detail about how he trained the model. Here’s how the books were filtered for pre-training: First, I downloaded the British Library dataset split of all 19th-century books. I filtered those down to books contemporaneous with the reign of Queen Victoria—which, unfortunately, cut out the novels of Jane Austen—and further filtered those down to a set of books with a optical character recognition (OCR) confidence of .65 or above, as listed in the metadata. This left me with 28,035 books, or roughly 2.93 billion tokens for pretraining data.
Getting it to behave like a conversational model was a lot harder. Trip started by trying to train on plays by Oscar Wilde and George Bernard Shaw, but found they didn’t provide enough pairs. Then he tried extracting dialogue pairs from the books themselves with poor results. The approach that worked was to have Claude Haiku and GPT-4o-mini generate synthetic conversation pairs for the supervised fine tuning, which solved the problem but sadly I think dilutes the “no training inputs from after 1899” claim from the original model card.
For the OpenClaw ecosystem, Mr. Chatterbox represents a compelling case study in how locally run, open-source models can be integrated into agent automation workflows. OpenClaw, as a platform for local-first AI assistants, thrives on plugins and models that prioritize data sovereignty and niche applications. The development of the llm-mrchatterbox plugin, built with Claude Code, mirrors the kind of tooling that OpenClaw users might leverage to extend their assistant capabilities. This project underscores the potential for historical or domain-specific models to serve as specialized agents within broader automation systems, even if their conversational abilities are currently limited.
The challenges faced in training Mr. Chatterbox—such as data scarcity and the need for synthetic fine-tuning—highlight the broader hurdles in creating effective local AI models without relying on modern, copyrighted data. In the OpenClaw context, this emphasizes the importance of community-driven datasets and open-source training frameworks. As more developers experiment with models like Mr. Chatterbox, the ecosystem could see a rise in plugins that cater to unique use cases, from historical research to creative writing, all running locally to ensure privacy and control.
Ultimately, Mr. Chatterbox is not just a weak Victorian-era model; it’s a testament to the growing movement toward decentralized, ethical AI. For OpenClaw users, it offers a blueprint for how to incorporate novel models into their workflows, fostering an environment where agents can be tailored to specific tasks without compromising on open-source values. As the platform evolves, projects like this will likely inspire further innovation in local AI assistants and plugin ecosystems.
