Nano vLLM

DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms

Noumena's internal version of your favorite coding tui but improved and optimized for our inference stack

Minimal Claude Code alternative. Single Python file, zero dependencies, ~250 lines.

Slop Scoring to Stop Slop

Unofficial CLI and MCP server for Lambda cloud GPU instances

PyTorch memory allocation visualizer

Reachy Mini Controllable Eyes mod. Or basis for any other hardware mod.

Accelerating MoE with IO and Tile-aware Optimizations

MoE training for Me and You and maybe other people

The best ChatGPT that $100 can buy.

Training framework with a goal to explore the frontier of sample efficiency of small language models

Speeding up the loading of ML models by efficiently splitting the weights across multiple disks then reading them in parallel during model initialization

Simple MPI implementation for prototyping or learning

[ICLR 2026] Learning to Reason without External Rewards

Shell completion using LLM for Qwen models

Tensor-Slayer : Manipulate weights and tensors of LLMs to achieve performance upgrades and introduce a novel inferenceless mechanistic interpretability

FULL Augment Code, Claude Code, Cluely, CodeBuddy, Comet, Cursor, Devin AI, Junie, Kiro, Leap.new, Lovable, Manus, NotionAI, Orchids.app, Perplexity, Poke, Qoder, Replit, Same.dev, Trae, Traycer AI…

Where GPUs get cooked 👩‍🍳🔥

All of the JSON data files and card images that the D2eCV utilizes for Descent 2nd Edition

Grabs sleep data from the Oura API, stores it in a local database and allows you to visualise the data with Grafana

Entropy Based Sampling and Parallel CoT Decoding

workaround for AirPod disconnection

StarCoder2-Instruct: Fully Transparent and Permissive Self-Alignment for Code Generation applied to llama 3 8b

[NeurIPS'24] SelfCodeAlign: Self-Alignment for Code Generation

A PyTorch native platform for training generative AI models