Skip to content
repository radar
Models & Inference

🧪autoresearch

ACTIVE BREAKOUT

karpathy/autoresearch

AI agents running research on single-GPU nanochat training automatically

⭐ Hugely popular: 94k stars, gaining about 563 a week

View on GitHub ↗

repo profile

vintage 2026 launched this year
language Python
license n/a

momentum

total stars 94k
stars added last week +563
HN peak 208 pts 5mo ago
commits / week 0 steady
issues closed 51% ~1 day to close, median
contributors 9
release cadence n/a
last activity 4mo ago

durability

backing community / independent
bus factor 1 solo
top-author share 68% 6 mo

bus factor = how many people it takes to cover more than half the commits (6 months). 1 is a solo project; higher means the work is spread across a team. top-author share is the single busiest author's slice of those commits.

since we covered it

monthly average + 11k/mo (+25%/mo) · + 51k total since PR#30

why it's a big deal

  • It turns model research into an agent-native loop, where the LLM is not just helping with code but actually running bounded empirical experiments end-to-end.
  • The design is unusually legible: one GPU, one editable file, one metric, and a fixed five-minute wall-clock budget make the whole system easy to reason about.
  • It points to a broader shift in OSS AI tooling from static copilots toward closed-loop systems that can test, compare, and improve themselves.

under the hood

  • The repo revolves around three files: prepare.py for fixed setup and utilities, train.py as the only mutable file, and program.md as the human-written instruction layer for the agent.
  • Every experiment runs for exactly five minutes and is measured on val_bpb, a vocab-size-independent validation metric designed to keep comparisons fair across architectural changes.
  • The codebase is deliberately self-contained and minimal, built around a simplified single-GPU nanochat training stack with no distributed setup or large config machinery.

our take from PR#30, 2026-03-25

star history

PR#30 · 43k94k now Mar 2026Aug 2026
  1. PR#30 43k 2026-03-25
  2. now 94k + 51k since first covered

curve is sampled from GitHub's star history, plus our own daily readings since we covered it; the dashed stretch is before we first covered it, the solid line since. figures at coverage are the numbers we printed then (approx.), current count is live.

understory

Better known than its recent output, coasting a little on attention.

-92 understory score output 0 · clout 92
Aug 2025 Jul 2026
  • output, commits & releases
  • clout, star velocity

output = commits & releases; clout = star velocity, both 0 to 100 monthly indices; the gap where output runs above clout is the understory. The understory →

covered in

  • PR#30 2026-03-25 on the radar

    Autonomous overnight experimentation for single-GPU model training

similar projects

compare these →
  • 🖥️ Open WebUI

    User-friendly AI Interface (Supports Ollama, OpenAI API,...)

    149k ACTIVE
  • 🦙 LLaMA-Factory

    Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

    74k ACTIVE
  • 💻 nanoGPT

    The simplest, fastest repository for training/finetuning medium-sized GPTs.

    62k ACTIVE

comments

Sign in with GitHub to add your blip on autoresearch.

loading comments…