Autonomous AI research swarm — runs ~100 overnight LLM training experiments unattended. Fork of karpathy/autoresearch with Digital Cognitive Labor routing.

Train your own ChatGPT on Apple Silicon — MLX port of nanochat

Train Llama 3 models from scratch. Any scale, any personality. By Arianna Method.

A minimal, hackable Vision-Language Model built on Karpathy’s nanochat — add image understanding and multimodal chat for under $200 in compute.

The best ChatGPT-style model that $100 of TPU time can buy.

A user's facts become a few rows in a content-addressed memory table — not a rewrite of the model. Paper, code, data, and interactive site for "User as Engram" (arXiv:2606.19172).

nanochat's inference engine re-vibed in C++ with GGML.

Ascend NPU fork of nanochat for LLM training with torch_npu/HCCL (experimental)

Erk — eCloud Tech. Türkçe LLM ailesi. nanosohbet: Erk modellerini sıfırdan eğiten tam açık reçete (Türkçe nanochat).

A legible and hackable port of Nanochat in JAX/Flax

I see now that the path I choose through the maze makes me what I am.

A 1.68B chat model trained from scratch (pretraining + SFT + GRPO) on the nanochat recipe. Base CORE 0.30 vs GPT-2 0.26, with a logit-level analysis of the RL checkpoint's regressions.

The official implementation of Ringmaster LMO, an asynchronous distributed optimizer for neural network training under heterogeneous compute environments.

Production-honest small language model training factory: data import, pretraining, SFT, eval gates, contamination checks, and GPU runbooks.

llama.cpp fork adding nanochat architecture support, so Karpathy's nanochat checkpoints run as GGUF. See NANOCHAT.md

Run nanochat training efficiently on Huawei Ascend NPUs with minimal code changes, supporting tokenizer, pretraining, and evaluation workflows.

The best GPT that $100-$125 worth of pre-training and finetuning can buy

I built this repo to prove to my granny that I can implement GPT.

The best ChatGPT that $100 can buy ported to an Nvidia RTX 5090 which is NOT $100

Autonomous ML research on a single DGX Spark: three agent modes spanning model capability (hypothesis generator to Karpathy's original design), custom GGUF support for frontier-scale quants like DeepSeek V4 Flash, a 4-level model/harness benchmark suite, and game-theoretic multi-branch orchestration