TensorLib is a pure-C deep learning library that trains tiny language models on CPU
Intelligent_Nose_791 · reddit · 2026-07-28
A developer built TensorLib, a pure-C deep learning library for CPU-only training, from scratch.
- It includes tensor manipulation, autograd, neural network modules, SGD and AdamW optimizers, and a decoder stack with layer norm, MHA, and FFN.
- The project also implements AVX2-accelerated matrix multiplication.
- Using it, the author trained a tiny language model with about 1.9M parameters on Tiny Shakespeare and shared sample generations.
- The repo is public on GitHub.
More from Infra
- Microsoft open-sources MarkItDown, a Python tool for turning documents into Markdown — mdancho84 · 2026-07-28
- Microsoft releases MarkItDown, an open-source document-to-Markdown converter — mdancho84 · 2026-07-28
- Screenshot shows Anthropic crawl spikes as users speculate Sonnet 6 training is underway — marclou · 2026-07-28
- NVIDIA says Jetson now fits in a bag while powering robots and edge AI — nordicinst · 2026-07-28
- India can assemble a finished chip in Gujarat, but the core is still imported — santoshpanda · 2026-07-28
- Open competition targets 36.8% faster Laguna XS 2.1 inference on consumer Macs — gajesh · 2026-07-28