# Torch Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/torch
> Markdown URL: https://aitinkerers.org/technologies/torch.md
> Technology record last updated: 2026-03-24T09:43:43Z
> Generated: 2026-09-21T04:43:58Z

The original open-source scientific computing framework and machine learning library (LuaJIT-based), directly preceding and influencing PyTorch.

Torch is the foundational open-source scientific computing framework, initially released in 2002, built on LuaJIT. It provides a powerful N-dimensional Tensor object and a C/CUDA backend for high-speed linear algebra and GPU-accelerated deep learning algorithms. Key organizations like Facebook AI Research (FAIR) and IBM utilized this framework for early-stage AI development. The project's core architecture, focused on maximum flexibility and speed, was later ported to Python, leading directly to the creation of the widely adopted PyTorch framework in 2017.

- Official technology site: https://torch.ch
- Public AI Tinkerers demos and talks: 3
- Result page: 1 of 1

## Recent Public Talks and Demos

### [Nanochat: training your own LLM chat bot from scratch](https://brussels.aitinkerers.org/talks/rsvp_dBCcFb9t3nk)

I'll walk the audience through Andrej Karpathy's latest reincarnation of minGPT/nanogpt, namely nanochat. This repo contains all major LLM stages including tokenization, pretraining, finetuning, evaluation, inference, and even a chat UI. Its minimalistic implementation allows you to train and evaluate an LLM from scratch, on a single GPU node or even a CPU. With access to better GPU machines (e.g. an 8X H100), you can get up to GPT2-level performance in under 2 hours!

- Event context: AI Tinkerers Antwerp Meetup - April 1 (no joke) — 2026-04-01 — Brussels
- Public talk page: https://brussels.aitinkerers.org/talks/rsvp_dBCcFb9t3nk

### [Reliable ML Predictions with Conformal Prediction](https://edmonton.aitinkerers.org/talks/rsvp_pjcUgYgII2c)

ML models make prediction but these predictions can be wrong. In high-stakes environments like healthcare, this has consequences. In normal everyday tasks like search or text generation, these wrong predictions are known as hallucinations. To integrate and augment ML models in society, we need them to be reliable! In this demo, I will show how to implement Conformal Prediction, a model-free framework that reasons how confident our model is. Ideally, if our model is not confident then, they should say "I don't know". However, current model deployments don't do this. Conformal Prediction wraps around any model to output a set instead of a single point. The sets are guaranteed to contain the true label with statistical validity. With this, we can statistically guarantee how reliable our predictions are instead of using adhoc unreliable methods that have no guarantees. I will be demo'ing how to convert your model to be more reliable with Conformal Prediction. This framework is general and useful in many applications. This will be a live demo that goes through the coding process and technical implementation of Conformal Prediction.

- Event context: AI Tinkerers Edmonton: March 2026 Meetup — 2026-03-25 — Edmonton
- Public talk page: https://edmonton.aitinkerers.org/talks/rsvp_pjcUgYgII2c

### [Hidden Torch](https://sf.aitinkerers.org/talks/rsvp_tjWLC0-9swQ)

Offline tutor running on a Raspberry Pi, building specific grade/location benchmarks, starting in Afghanistan.

- Event context: AI Tinkerers - San Francisco - January 2025 Meetup — 2025-01-30 — San Francisco
- Public talk page: https://sf.aitinkerers.org/talks/rsvp_tjWLC0-9swQ

## Related Technologies

- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 662 public demos
- [Conformal Prediction](https://aitinkerers.org/technologies/conformal-prediction) ([Markdown](https://aitinkerers.org/technologies/conformal-prediction.md)) — 1 public demo
- [Edge AI](https://aitinkerers.org/technologies/edge-ai) ([Markdown](https://aitinkerers.org/technologies/edge-ai.md)) — 5 public demos
- [Jupyter](https://aitinkerers.org/technologies/jupyter) ([Markdown](https://aitinkerers.org/technologies/jupyter.md)) — 8 public demos
- [Large Language Model](https://aitinkerers.org/technologies/large-language-model) ([Markdown](https://aitinkerers.org/technologies/large-language-model.md)) — 9 public demos
- [LLM](https://aitinkerers.org/technologies/llm) ([Markdown](https://aitinkerers.org/technologies/llm.md)) — 123 public demos
- [On-device AI](https://aitinkerers.org/technologies/on-device-ai) ([Markdown](https://aitinkerers.org/technologies/on-device-ai.md)) — 3 public demos
- [Raspberry Pi](https://aitinkerers.org/technologies/raspberry-pi) ([Markdown](https://aitinkerers.org/technologies/raspberry-pi.md)) — 8 public demos
- [Transformers](https://aitinkerers.org/technologies/transformers) ([Markdown](https://aitinkerers.org/technologies/transformers.md)) — 148 public demos
- [Wandb](https://aitinkerers.org/technologies/wandb) ([Markdown](https://aitinkerers.org/technologies/wandb.md)) — 1 public demo
- [Weights & Biases](https://aitinkerers.org/technologies/weights-biases) ([Markdown](https://aitinkerers.org/technologies/weights-biases.md)) — 9 public demos