Experimental fork of Karpathy's nanoGPT modifying the architecture and training loop to make per-source influence over predictions identifiable and controllable.

中文小黑怪诞正文配图生成 Skill | 16:9 白底手绘 | 少量红橙蓝批注 | Codex Skill

[arXiv 2026] Learning from Rare Success and Rich Feedback via Reflection-Enhanced Self-Distillation

Bird is a cli for twitter, so your agents can tweet.

LLM-compiled knowledge bases for any AI agent. Parallel multi-agent research, thesis-driven investigation, source ingestion, wiki compilation, querying, and artifact generation.

🧻 Knowledge system for agents. Local-first, file-based, progressively disclosed.

Compile programs directly into transformer weights. Includes a 2D convex-hull KV cache with O(log n) inference.

Interactive version of the CuTe layout paper

A pure-Python implementation of the Nvidia CuTe layout algebra intended to be approachable and easy to learn.

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

A markdown native slides tool for academics building with agents.

Statistical analysis methods for comparing prompt and model performance in LLM evaluations.

FlashSampling: Fast and Memory-Efficient Exact Sampling (https://huggingface.co/papers/2603.15854)

A demo of what it takes to get Kimi K2 Thinking to train on top of HuggingFace

100M tokens. Infinite compute. Lowest val loss wins.

Minimalist digital clock with multiple timezone support and JSON configuration. Clean and customizable interface.

NERV Operations Console aesthetic for web interfaces — Evangelion-inspired instrumentation UI skill for Claude Code / OpenClaw

AI agents running research on single-GPU nanochat training automatically

Sparsity as a Variance Regulator for Improved Depth Utilization in Language Models

The agent that grows with you

Sutskever 30 implementations inspired by https://papercode.vercel.app/ | For Agents, use https://github.com/pageman/Sutskever-Agent | Polyglot / Multi-Backed version at https://github.com/pageman/s…

Educational WIP

we have ai at home

TEVO: evolve LM motifs cheaply, then validate them in downstream train.py loops.

Dialectical reasoning architecture for LLMs (Thesis → Antithesis → Synthesis)

General plug-and-play inference library for Recursive Language Models (RLMs), supporting various sandboxes.