3 GitHub repos worth reading for the code: nanoGPT, tinygrad, llama.cpp
cneuralnetwork · x · 2026-11-12
A curated list of three GitHub repos recommended for reading the source code rather than just using them:
- nanoGPT: understand how GPT works from scratch, including attention, transformer blocks, and token generation.
- tinygrad (tensor.py): learn tensor operations, autograd, and how a deep learning framework is built.
- llama.cpp (src/llama-model.cpp): see how LLM architectures, model loading, and computation graphs are handled in C++.
The author also asks for other repos that belong on the list — a handy learning path for developers who want to go below the abstraction layer.
More from coding & agent
- One prompt, Blender MCP: Qwen4-Max builds a ramen stall at ~1/4 of DeepSeek's token cost — teortaxesTex · 2026-11-12
- 1.7B model shown driving browser and writing HTML autonomously, claimed to hit 4o-level benchmarks — JosephJacks_ · 2026-11-12
- Hackathon Project Uses Hermes Agent to Turn Prompts into 3D-Printable CAD Designs — alexcovo_eth · 2026-11-12
- Baidu's ERNIE 5.5 Flash writes its own code to build editable CAD skyscraper model — heyshrutimishra · 2026-11-12
- Gamma's UGC pages become parasite SEO target, risking domain-wide Google penalties — gefei55 · 2026-11-12
- A neat prompt to make your agent expose its hidden assumptions about you — keyanzhang · 2026-11-12