RobinKa
Follow
Stars
🚀🚀 「大模型」2小时完全从0训练26M的小参数GPT！🌏 Train a 26M-parameter GPT from scratch in just 2h!
A modular, primitive-first, python-first PyTorch library for Reinforcement Learning.
Official Jax Implementation of MD4 Masked Diffusion Models
A Datacenter Scale Distributed Inference Serving Framework
☁️ Nextcloud server, a safe home for all your data
Nvdiffrast - Modular Primitives for High-Performance Differentiable Rendering
High-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently.
Apache Superset is a Data Visualization and Data Exploration Platform
TensorZero creates a feedback loop for optimizing LLM applications — turning production data into smarter, faster, and cheaper models.
Witness the aha moment of VLM with less than $3.
Finetune Llama 3.3, DeepSeek-R1, Gemma 3 & Reasoning LLMs 2x faster with 70% less memory! 🦥
🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
NVIDIA Linux open GPU with P2P support
You like pytorch? You like micrograd? You love tinygrad! ❤️
Fully open reproduction of DeepSeek-R1
Official repository for our work on micro-budget training of large-scale diffusion models.
TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching
AlphaFold Meets Flow Matching for Generating Protein Ensembles
A Python library transfers PyTorch tensors between CPU and NVMe
Statsmodels: statistical modeling and econometrics in Python
A PyTorch repo for data loading and utilities to be shared by the PyTorch domain libraries.
Quantized Attention that achieves speedups of 2.1-3.1x and 2.7-5.1x compared to FlashAttention2 and xformers, respectively, without lossing end-to-end metrics across various models.